 Let's get started. I'm going to hand over in a minute to Roderick Sadler from La Trobe University. With him are Charlie Ambrose and Simon Huggert. They're a great team and they've done some really interesting work that Roderick's going to tell you about. So I'm just going to hand over now to Roderick Sadler from La Trobe University. Thank you very much Simon. We've got this this Claiming the Territory title and that's come out of a comment I made in our recent video presentation that you put together. So I guess it's led me to think about lots of different aspects of claiming the territory that have been part of this project. I've actually prepared material looking at one particular aspect of that. I suppose what the comment originally referred to was was the notion that putting research data material out and making it available is actually it's actually an assertive act. It's an act of claiming a territory and we've certainly come across a lot of researchers who are concerned that that it puts them in a vulnerable position that it makes them that's revealing their material and perhaps compromising their ability to to exploit it ahead of other people who may take an opportunity. So I guess we're still working on that idea. How do we turn that into a sellable proposition and how do we give researchers the tools to be confident with taking that sort of step? Certainly it's still a comment that we're hearing from quite senior academics that that there are concerns that exposing your material before you've actually worked at thoroughly is something that they have concerns about. So it's not something that we can say we've dealt with. It's something we're still actively working on. But the notion of claiming the territory lead me to a lot of other thoughts. I mean we're dealing with Trove here we're dealing with our metadata stores projects as a library a library based project. I mean the other projects I know have been steered from other parts of their institutions. But to me it's it's quite a it is a claiming the territory exercise for a library to be doing this kind of work. As we've been going through the projects so many aspects but are really reworking off of traditional library tasks using traditional library based skills. It's basically in a sense it's a new form of librarianship. So it's kind of territory that we should be claiming. So management of identities describing materials all that sort of thing. It's library work from time immemorial. What I wanted to focus on was what I suppose emerged out of this project as the big sleeper. It's the it's the thing that we probably didn't see coming from an early point but it's turned out to be more and more of an issue I suppose which is the the researcher identification management. Now I guess we're at a time at the moment with the rise of social networking and so forth where there are so many avenues for for people to for the identification of people to either through their own efforts or through the efforts of other organizations and so forth to become described in one way or another and and those seem to have become sort of an uncoordinated mess in some respects. I'll just go through some screenshots to give some example of the territory that we're looking at and to start thinking about how we could claim it I guess. So I've just clicked onto a page. It's just a screenshot from Thompson's researcher ID.com so we've got a record here for a Latrobe academic quite a senior academic and he's done work on his own behalf to get himself a researcher ID allocated obviously through Thompson to a commercial company. Now Latrobe has had pretty minimal take up from academics for for this service but this is one this is one academic who has taken this step so he's built himself a profile and he's allocated a number of publications obviously that couldn't fit his publication list into a screenshot but so he's he's one who's got himself a researcher ID by Thompson's. The next next screenshot is a Latrobe researcher who's built himself a research gate profile so again his own initiative he's done something I suppose comparable to what the previous one did through Thompson's researcher ID so he's he's put together a profile of himself again of course the publication list can't fit in the screenshot so we've we've gone from Thompson's to research gate now totally different sort of enterprise Thompson's being obviously they're making researcher ID.com a freely available available service still coming out of from quite a different sector. Then we move on to another Latrobe academic who's built himself a profile through Microsoft Academic Search so he's done quite a bit of work put it in a very fetching photo of himself and um he's built quite a profile in the system for himself. Now I think Simon sorry I'm just going to talk a little off the site here so I'm hugged um does he have you spoken to him about using this as a primary sort of profile page? I've talked to him about what he's he's used and he said he he tried a few different sort of profile sites and sort of some of them you can see I looked at a number of them and you could see some of them he went in there and did something at one point and then obviously it didn't sort of meet his needs and he's only done a couple of things added a few things and then gone on to something else and I think there's a couple of ones that he is using I think Google scholar and Microsoft academic are the ones he's actually I think actively looking at and I think the other thing about those profile pages is that they also they work on you they send you emails they say someone looked at your profile someone commented on this there's a new publication that might be yours here you might want to claim that all these sort of things and so there's there's some that are quite active and some of it that you don't you know not much happens unless you actually work on them so he he tried them out and was interested in you know sort of publicizing what he's done and putting a bit of a profile out there so there's a couple of years actively using. Okay so moving on from Hilton this is another low trade academic no no photo at this time but he's he's actually built himself a Google scholar profile so he's again put together citations etc he's brought them together under one identity and another one in in Google scholar he's been quite active in building his profile here so he's he's taken the trouble putting a photo in and so forth so for each of these these these academics are acquiring ids in different systems but but these are all these are all uncoordinated and unmanaged so the same one sorry I'll just go back again so so Russell Huy sorry I just so Russell Huy we just looked at through Google scholar we'll start looking at him now through a different structure of of identification so so the ISNI identifier so Russell Huy has an ISNI identifier so the international standard name modernification system which is a quite a new initiative but it's a it's a very high level initiative so so it's this particular interface we're looking at is powered by OCLC in the library world you don't get a lot bigger than OCLC uh an ISNI is based in has an international standard as its basis so so in contrast to the the sort of commercial efforts with through Thomson's and the community-based ones we now have a very high level organized sort of approach to to identification management so we've got an ISNI record here for Russell Huy and the same researcher again so now this is the VIF a VIF record for Russell Huy so this is the virtual international authority file so we've I said in a sense we've moved down a level so and I'll just click through we've got some more detail so so this is back to the to some detail from the ISNI record so the ISNI record is actually sourced from the VIF record so I'll just click back a couple of slides so so the ISNI record here with its various details is is generated through that from that VIF record so we've got this sort of hierarchy ISNI kind of locating itself at the broadest level um so there's some other sources they are listed for the for the VIF record as well so VIF is highlighted one here um so now the VIF record in turn is sourced from a record from the National Library of Australia so over on the right we've got a little diagram there of this sort of structure of sources for this record and on the left we've got we've got them listed um so this this VIF record for for Russell Huy has has been put together from a number of different sources and it's just starting to sound familiar when you've done anything with the National Library's party infrastructure so it's a kind of um a conglomeration sort of effect so the VIF record um this is the record history these are obviously screenshots attached to the same record so there are there's quite a lengthy list of record IDs there some of them occur several times because there's some deletions and so forth but but that VIF record has has been put together out of out of work over quite a number of of different records from different sources so there's an awful lot of work going on out there in this kind of um this identification management a lot of different agencies a lot of them working at a national and international level and they're actually trying to trying to coordinate their efforts which is something that's quite a novelty for someone who's been working in in the library world for some time to see all of these agencies coordinate is quite extraordinary so from the VIF record so uh as we saw the VIF record had a um had the National Library of Australia as I've highlighted there National Library of Australia is one of the sources for the VIF record so fix through that one so this is the the National Library of Australia people system display for that um for that record so so Russell Hoy again so he's making this is really following him all the way down the hierarchy I suppose from from ISNI through VIF down through to the National Library so the um the TRIVE record lists three identifiers for this person I suppose whether you describe it as an identifier for a person or for a record is a moot point but anyway um so we've got two or two libraries Australia identifiers in this case which is a little unusual I think he was described twice erroneously anyway and um anyone who's familiar with the party infrastructure will see how the the familiar form of the NLA dot party identifier at the bottom of that list of identifiers so so he's found his way into the party infrastructure out of the libraries Australia infrastructure so um Andy's found his way right up to the ISNI infrastructure through VIF as well so when we look at the libraries Australia record now it takes another twist um the libraries Australia authority record is originally a library of congress record so we've got this extraordinary twisting and turning going on with these identification of this one person um so um the LC number field there is indicating library of congress number so if we trace that one back so library of congress has a record for the same person so so we've got Russell Hoid described in many various places but um some some linking some working independently so we can see this there's this interconnection going on um it's not always um two-way the the library of congress record for instance doesn't have any reference at all to the um libraries Australia record the libraries Australia record refers back to this one and then the national library party infrastructure record refers back to the um to the libraries Australia record but the libraries Australian one doesn't refer the other way to the parties record so now just to to add to this kind of proliferation of identifiers from all these different sectors we have our own local version of a party record so this is the La Trove approach we've got um what we're doing here is I'm just looking at the lower part of the screen we've got um I'll just look at the external ones first so we've got a Trove record link so so this record is aligned with a Trove record so we've included that now this is this is a public record in our in our repository system so this is part of this is actually part of our meta metadata store infrastructure as we've realized it through our um our La Trove repository so we've actually modified the display so that it takes a trove link and um puts it in the public display with a link we can incorporate a VIFF record link as well so it can go to the la level it can jump up to the VIFF level then off to the side we have a scopus record ID scopus obviously being a commercial organization so sitting outside that structure but yet we we've taken the decision to incorporate all of these IDs in the same record and at the bottom we have an ISNI identifier and as far as I know ISNI identifiers are not um currently they cannot currently be resolved as URLs but they are intended to be um to be unique identification um for researchers in the world so something like an ISBN for a book where it's not something you can resolve to URL but it is intended to be to be unique so so I'd have all of that um obviously the the the new one to the mix is our local record ID so now we we're working with red box and mint in the back end so um the records we have in here need to coordinate with our with our mint records when it comes to identifying researchers so so um the identifier there for anyone familiar with mint is probably not going to look familiar as a mint identifier usually a mint identifier will be um uh along um a hash hash key so it's it's derived out of something um deep within the technical workings of the mint system put together intended to be intended to become something unique but but to us it was it was um it wasn't something it was going to help pull our systems together so we wanted to to push the records out into this public interface um so what we've done with our local ID is we've we've taken um an identification number from from La Trobe's um research management system to our research master system and we've pre prepended the La Trobe isle code to that so the au hyphen vlu is what we've prepended to that to um to make our records specific to La Trobe so obviously if we're pushing if we're pushing a party record out to say national library and we just put a research master team digit numerical code there perfectly possible that could that could be the same as another one someone else contributed so we've we've taken this step of um of trying to ensure uniqueness by by putting identification in that form so um that that's form of um identify is the one that's going to be the key that we send up to um to the national library in uh as a so we we actually uh contributing eaccpf records to the national library and we're using our identifiers of that form as our primary match point so um so the the idea of that is that we can at any at any point where we see one of our records we can we can readily take it back to the um original source so so we just take off the isle code and we have the um the research master identifier uh now the thing about the research master identifier is that it actually enables us to coordinate party records with um not only our research data records where we're explicitly adding the identifiers but um but also the publication records coming out of our our research master system so they all come with um with uh research identifiers as well so this is actually building a system where we can pull together this whole kind of um this whole ecosystem of of of data about the tropes research output so by by building our research identification around this this number it actually enables us to coordinate identification of publication records which are also in our repository system with um with research collection records which will be here as well and and to push those um those records out to national library maintain those those linkages so um we we haven't gotten a lot further at this stage than then I guess incorporating all of these these records I mean there's still some open questions about um the proper kind of mark-up for some of these identification of those identifiers like a scopus id and so forth I mean it looks like it's just a pretty vanilla um field that you include in the record at this stage rather than being formally identified but anyway this is the kind of work that we're attempting to to do we're trying to bring together these identifiers now um this particular researcher didn't have a uh a researcher ID identifier we looked at one previously who did have but this one doesn't have otherwise we would have incorporated that as well this one doesn't have an orchid ID either so I guess there are a lot of questions there about are we going to continue trying to keep all of these balls in the air I suppose it's the you know the kind of territory we're looking at if we come back to the notion of claiming the territory we are looking at researchers who are in some cases quite active in in managing their own profiles they're out there on these other systems like google scholar thompson's researcher ID research gate various others they're managing their own identities out there so I suppose the sort of thinking we've had here is that we have to engage with that or we're just going to be out in the picture so we we're not in a position to to um require them to to restrict their activities to one system or another or to impose another system on them so we're trying to make our work complementary to what they're already engaged with I guess that's is that you say that's a reasonable way of putting this on yeah yeah so we we're trying to we're trying to engage with the the landscape that our researchers are engaged with so um so pulling those identifiers into our system hopefully is a way of making our our records relevant to the sort of things that they're doing and these various other initiatives and I guess ultimately when these records are pushed to RDA to research data Australia it will also push those identifiers out there and and you know the data is out there in the public domain so if that researcher wants to yeah promote them their work and claim their work claim the territory in different systems they can do that and we're integrating with that as much as possible so I guess there's still there seem to be some issues about when you know when um party records are pushed from NLA to RDA or back again or I mean there seem to be there seems to be data that gets lost um somehow in transit that's something we're still working out for ourselves like what's the what's the approach that we can take that will in it will keep the best integrity for the the sort of work we're doing so we don't want to be putting putting various elements in our records and find that they disappear when they get new from one system to another so we're still working that out we haven't come up to you know we haven't finalized our approach to some of these things yet but this is kind of the kind of direction we're heading I guess it's fair to say now um okay so we're looking at something specific to do with our research master to red box slash mint to our repository to the NLA to that connection so we've got some slides about how our metadata stores being configured to to keep this researcher identity consistent as it goes through each component of the system so just just um recapitulating there that our ISO code for the probe is our I guess that's part of claiming our territory too we're sticking our brand on our research ID and so we're we're associating that with our our the research ID that the trobes are familiar with so um excuse me a minute Roderick yes Roderick just excuse me a minute somebody's got their microphone unmuted so we can hear conversation in the background can you can everybody mute themselves okay let's try that now I still hear conversation that might have been us I will try I'm not sure okay away you go sorry okay thanks Simon so um so we just we're just looking in a bit more detail at at how we put together our local identifiers and how we push them through our systems so so we've got our ISO code there and we we concatenate it with with a an institutional ID for the for the researcher so we've got our um we've got um CSV file that we've we've pulled out of research master so we've got highlighted in yellow just about the middle there um Peter Tate and um so we've at this stage we've just got the on the left hand side we've just got the the raw institutional identifier for this researcher so that's that's unique in the latrobe context but obviously not not necessarily beyond the trobe so as we move through um we go through to to the red box mint view what's is the mint view obviously um so so towards the bottom of the page we've got boxed in red there um several instances several instances of um of our reworked version of that identifier so so we've actually we've got the au dash vlu um iso code for the trobe prepended now so whereas previously we had the previously we just had the 10 digit code with the the five leading zeros now we have the iso code and and that same identifier so this is what we we're going to be using as our primary identifier for this this research so um so there are various other identifiers that that um red box creates and invariably they involve an extremely long hash key so so our decision is it's it's it's more readable it's more portable it makes sense across our different systems to to work with a different identifier so just passing over charlie i'm going to just add to that and say that that um what appears here is no and id's would previously have looked something like what we've got for the storage id it looks like a hash value okay so so this is the mint view so we've gone from the research master csv extract through to the mint view and um so now we've just got a view of the um uh configuration that what we did in the mint or charlie really did in the mint um to make the change to get the to get the um our defined identifier as the as the key one in this process so um the standard mint behave if your record gets pushed up to an LA is to is to look for what it thinks as what it thinks of as the identifier for the for the researcher so that would be as as charlie said that would typically have been something along one of those values towards the top of the record that's what mint would have expected to have been to have been sent to an LA now now in our model um we've we've actually uh worked around I suppose the the mint standard processes in the sense that we've we've actually we haven't pushed party records from mint into our repository system now we looked at a repository system um party record before that's actually populated from a parallel process so so when the records are put together out of out of research master they go in two different directions they go they go into um into the vital um the repository system and they in parallel they go into the mint system but they they have the same identifier so that's our why pulling them together so the oai harvest from nla is actually out of our our vital fedora repository so um when we come to use an identifier out of mint to use party record and it it doesn't have an nla identifier attached to it it will do its process of of going up to nla to look for one so to look for its identity the identifier that it's configured to search for through an sru query and pull back an nla identifier so so what we've done um using this code is to um is to change what's stored for what's searched for so we've we've we've changed the um the value that will go into the sru query so it will go off looking for one of those au dash vru 000 etc identifies which is what we we will have we will have pushed that up to nla out of our vital um repository using the oai harvest and that's what mint will mint will coordinate with that to pull back the nla identifier now that it sounds like a sort of roundabout process but it's actually worked quite smoothly for us we've um we've actually I suppose quarantined the use that we've made of mint um we haven't led we haven't let mint and red box lead what we're doing here we've we've actually I suppose configured them to fit into the processes that we wanted to run so um but we've actually managed to get them to fit in quite well with a with the approach we've taken so so we've got oai harvesting concentrated in our repository system so we're already we're already using that system for other other oai harvesting now we're extending it to use it for um eac cpf and for rift cs and so far that's actually worked very smoothly for us so that coordination between between the oai harvesting out of the repository and the um the su lookups and um and the bringing back of the nla identifiers into mint has actually worked very smoothly so we've been quite happy with that with that result um but it did take some system configuration work to to get that happening so credit to charlie basically for sorting that out so I guess that's one of the ways that we've we've differentiated our our implementation of red box and mint is that we've we've actually um got it coordinating with a bigger process I suppose we haven't taken the out of the box sort of processes that it runs with we've we've come up with our own process of processes and we've we've configured red box and mint to work with that so I guess that's the essence of um of how we're keeping our identifiers um in a form that um it's meaningful to us and not just in the red box mint system so I guess that's um that's the essence of um of what I have prepared in terms of slides so forth now simon did you want to pick up and say anything further um thanks ruby I guess just to to wrap up so that's all right so it's it's simon hug it here um I'll just say a couple of things and then we'll hand over to simon pockley to you know arrange to mention about questions and things and how we're going for time but I guess uh the thing we have to sort out here is to as to how what level do we manage those identities and all of the the different external systems that people are using so we had to we have to go in and edit those records party records for people to add those identifiers in and then because we do that work it gets pushed out to the systems that we want that to be exposed in and therefore you know we have to look at how much work it's going to be to actually manage all of that stuff and how we resource uh you know managing those identifiers but the university sees that as a really important thing that we should be doing so I'm hoping that that means they'll provide some resources and support for us to be able to do that at some point so you know this is at this stage we're looking at that and demonstrating that's what we can do and then we we figure out if we can manage that across you know 5 000 researchers at the trove or however the number of people we have so Charlie did you want to say anything we're done okay so I think we're done Simon and do we want to I'll hand over to you for asking questions well uh Simon thank you and Roderick thank you for that Charlie as well um I I'll start with a question well can I just say if you haven't got a microphone please use the chat box for any questions you have um if you have seen the video that we made a week or so ago uh I found it extraordinary because we had a quite a lot of of material there that I had to edit down and one of the things that Simon Huggert said was that one of the core challenges that they had at La Trobe with their metadata stores project was actually getting the collections from researchers getting the the researchers engaged what Roderick's been describing is very much the sort of under the bonnet side of building researcher profiles and I wonder if you could talk a little bit about how researcher profiles fit into the La Trobe vision and just how important are they as a driver for that engagement with researchers well I've just been generously handed the microphone from Roderick um so Simon Huggert here again the the researcher profile sort of things is still reasonably undeveloped underdeveloped at La Trobe um so we're in a very good position I think to be able to engage with researchers and say if we manage their identities manage their publications manage their research data collections and the research data that they they want us to curate and make available then we can provide researcher profiles either from directly from our repository which is what we're able to demonstrate now but it's it needs some styling and some work to to improve the look of it or pushing that out to other systems at La Trobe to be able to you know allow researchers to to manage their own identities in some ways and and profiles and and you know use other systems that sort of do that that work a lot better so La Trobe has a has just gone through a a content a web content management system implementation that took a couple of years and part of that implementation while is getting researcher profiles out there on the website so that researchers themselves can log in and you know create a description of themselves put their publications out there and describe what they're doing and their research areas but at the moment that's a that's a completely manual process so each researcher can go in edit their page they have to manually cut and paste all of their publications into that page and save it so it's extremely inefficient so we're seeing the work we're doing here as a way of saying look we can demonstrate how some of this work can be done from or a lot of it can be done from the repository and data pushed out to the right places and managed to to help researchers with their profiles so I guess what what you've seen today is probably you know the the bare bones of it you can see that we've we can demonstrate we can pull in you know profiles from from different areas and different external systems we see that as a key you know selling point to researchers because that they're obviously engaged with what what you know the different systems are out there and what they can provide for them and and there's a and if we can say well we're in this space and we can manage some of that for you that's a big a big boom for academics because they their workloads are very high they they have little time to manage this kind of stuff and we we're at the moment providing resources and being able to do that coming for them so they so researchers do they do get excited about you know when we talk about profiles and being able to manage things for them but we you know we're still in a process of figuring out how we might do that thank you if anyone has it with a mic has questions please unmute yourselves and speaker otherwise use the chat unless anyone well in the meantime I have another question for Roderick and this is really it's another claiming the territory question and it's related to something you said when we did the video about the change in the library as it becomes the first publisher can you expand a bit on that the sort of changes you're seeing that the library has to make as a first publisher well it's um it's it's an interesting thing it's almost the it's a bit of a back to the future kind of situation because at La Trobe with the with the cataloging work that's down here with regard to the regular collection the emphasis was going away from working with identity with authority records it's going away from that kind of that control sort of work and just simplifying more and more and assuming that everything would just fall into place so what's happened now with this project is that suddenly authority works back in the frame again it's back as a a central task so managing these records identifying people this unambiguous identification of researchers is now back as a central part of the of the discipline of what we're doing so it's this and it's because we're in that role now so we we can't assume that everything we're dealing with has been already dealt with by someone else so if you're doing normal cataloging work you'd assume that anything you're cataloging anything you're handling has probably been cataloged by the Library of Congress or National Library of Australia or or whatever so so you're down to a very minimal task um but when you start dealing with say research data collections and um you're dealing with uh you obviously need to associate the correct researcher with the correct you know the collection etc so now we're back to uh in a sense we're back to that kind of um high-level cataloging work that was was slipping out of favor in the library in the library context but we need to um we need to recover those sort of those sort of skills with along with skills with dealing with um in some cases quite complex digital content and uh engaging with quite complex research work so we need to find um I guess we need to find the language for dealing with all that kind of thing it's not just a matter of dealing with a different different metadata schemes or that sort of thing it's also finding the I guess it's finding a way of actually talking to the researchers because we we're not going to be building um decent collection descriptions if we're not talking to the researchers if you have a book in hand that you're cataloging more often than not you will just you will just um do it yourself but if you're dealing with a collection um when you're the publisher you're the one who has to make the decisions about the kind of structure you put around it like I mean there are there are very fundamental questions like what is your collection at the moment we've we've um been given a lot of material from uh from a uh research center which has quite recently wound up but it it had a lot of valuable research material in it so it's a it's a pretty fundamental question like what do you call your collection is the entire thing a collection that's of pretty limited value um I mean it might be value it might be a value to have one description at that level but then it's been a question that um people in libraries have had to ask for a long time you know where where do you actually target your description so I think for that collection we we're going to have to to look at some multiple collections to to give more meaningful descriptions because no one else is going to be doing it we're the ones who are going to be doing it and if we don't if we don't do it well no one else is going to be doing a better job of it so if we mess it up then um you know we've messed it up for the world basically we've we've let down those researchers so I'll pass on the song he wants to say I think just also one of the things is the researchers themselves have a big ownership around their data that they want to publish and so they do you know they're very keen for us to put it somewhere put a doi around it make it available publish it so that's that that claiming the territory and being able to sort of publish things for the first time but they do you know they they get worried about how how their data is exposed and described and the rest of it and who's going to get access to it which is you know quite rightly so so it is as what we're saying it's important to get that right and make it available but they are they are very engaged with that because uh we're seeing I think with some discussions with people there there are some journals which are asking for the data as part of the review process of um when a paper is submitted for publication so the data is made available so that the reviewer can look at that and verify the the results in the paper itself and therefore accept it for publication so if the library can be within that space as well we're engaging with the researcher as soon as they publish or when they're about to publish or just before they publish we're getting their data at the same time we're providing value in that process and that's that's really important to as a part of engaging with our researchers and what they're doing are there any skills um the the data librarians don't have that you need to train them up in when it comes to data um that's a that's a very good question we have a kind of an interesting mix of people in our group at the moment that have you know really good skills around analyzing you know data itself and exposing it and putting it making it available on it systems that kind of thing so it's probably the skills that we would need to to beef up in uh you know a metadata librarian that would be dealing with the data and looking at it and being and having to describe it and make it available so but I I see it as a kind of team approach I wouldn't say say that you have to have one person that has all of those skills it really does you know is a matter of the team looking at things and and the people who are specialists in certain areas providing advice and and help in that regard when I think there's there's too much sort of you know in our experience so far there's already too much diverse data and that we're seeing from different projects for any one of us to to understand it all um and and by having a group of people who sort of have a mix of uh you know librarian skills just description and dealing with information and information management as well as you know file management and and data and data sources and then you know servers and systems and making things available you you know you're not going to have one person that's going to have that having a team that can cover all of that is really important I think Ja Ben has a question Simon I'd like to know your metadata store a project as a project itself or a finished song so do you want to talk about how to make it a long-term sustainable electoral yep so how are we going to make the metadata store sustainable as a long-term thing at the trove um so in terms of uh so we have the processes in place now to be able to manage the metadata store so that the thing that the the issue that's coming up now which I guess is common across these projects is you know how do we how do we resource getting people out there and talking to researchers to to find out what the research they're doing what collections they have how they want to expose their data and make it available in the metadata store so there's how do we sustain that resourcing and then there's the issue of having the staff at our end to deal with what we're being given what's coming into our repository so that's a good question that we're going to have to manage the approach at the trove is to where we're going with let me let me think about this for a second we're dealing we're trying to I guess work with the research services office the research office at the trove and the office of e-research so a collaborative approach with them and to some degree with ICT to to kind of share that load about making contact with researchers dealing with their data dealing with you know the questions that they have and making the connection so so that we're not doing all the work so if someone is having a conversation with a researcher then they can refer that them to us but that's that's kind of the the sort of research data management sort of area and systems that kind of thing but we're also now working with our liaison librarians to get them to be our advocates and and talk about because they're they're dealing with researchers anyway in terms of you know literature searches grant support you know all sorts of areas through the you know publishing and the research lifecycle but and so we want we want to train them up and have them understand what are the issues that we want to find out from researchers so that we can help them and get their data exposed so they're also our biggest asset I guess in terms of numbers and and so we're going embarking on a training program with them at the moment and we're also we're also looking at position descriptions and what are the types of people we need are there dedicated people that are going to sort of be within a liaison librarian team that do this kind of work or is it incorporated into the day-to-day work that they do there's probably a mix of models out there as to how we're going to approach that and we will also be looking at our our team within our digital infrastructure team that manage their repository and looking at our resourcing and roles and you know what dedicated resources do we need here to be able to manage this work so an example of you know a set of research data that we've got it's it's a huge storehouse of data that we've got from a previous center that closed down and it's that work could potentially you know go on for years and we've already spent you know I guess four or five months looking at that data and we know we're only just at the point of we haven't even described it as part of this project yet because it's not not able to be publicly exposed yet but you know just analyzing what's in that collection and providing advice to there's a steering committee involved with that is a big part you know and it's just one collection so resourcing and sustainability are big issues but the we have our ends in a part of our project we have our steering committee and we're going to re-engineer that steering committee to be probably a subcommittee of our research and graduate studies committee one of our official committees at the university that will then oversee the project from then on and and that will involve as many stakeholders as we can from ICT from e-research research and from faculties and departments themselves and the library so that hopefully will keep people well informed of what work we're doing and getting the word out there but also the need for resourcing and management of this as a you know a whole service for the university thank you very much that's probably unless there are any other questions it's probably not a bad note to end on I recommend if you haven't seen it that you have a look at the short video that we've made on the La Trobe project with the team so that you can see who's who and I'm sure that Simon Roderick and Charlie would be happy to answer any questions if you've got them by email or phone or whatever so at this point we might wrap it up and I'd just like to thank you Roderick Simon and Charlie for sharing this your approach to identification I think has interested a lot of people and particularly entering into that discussion about the changes that are afforded so thank you very much