 Good afternoon everybody. My name's Jerry Ryder. I work with the ARDC, the Australian Research Data Commons, and I'm based in Adelaide. It's my pleasure to welcome you all here today to our webinar, where we'll learn about some of the latest updates to our online services and hear from the ARDC DevOps Manager, Joel Ben, who I'm sure many of you are familiar with. So that's probably enough from me. I'd like to hand over now to Joel, who will take us through the recent changes to ANZ Online Services. Well, thanks everyone for tuning in. My name's Joel Ben. If you haven't met me, I'm the DevOps Manager for software development within ANZ. ARDC, I should say. And I'm going to walk through today some of the changes that we've put in for Release 27 and 28. So things to be covered today. I'm going to talk a little bit about the DIY service changes that were implemented in R28, a new feature in Research Data Australia, which is a relationship graph visualisation. As Jerry has mentioned already, this is the International Geosample Number Service that was developed. And I'll go through the RVA, Research Recabularies Australia notifications that we put in Release 28. Sorry, I'm on a different computer and the mouse scrolls in the complete opposite way. So I'm bound to trip over a few times. I'll start with the DIY changes. So as part of R28, we put in a few changes which were instigated by Data Site and ANZ. We put out an email for comments to all DIY clients to feedback if there was any impacts on them. And one of the changes that we proposed was the removal of client identifiers from DIY streams. So there's a little example up here where you can see the top DIY. It's got 16, which is a client identifier. And then the DIY is how it will look now without those identifiers. So in the early days, when we initiated the DIY system within ANZ, we came up with the notion of putting client identifiers within the DIY streams so that we could identify whose DIYs were whose. And also, in a future enhancement, we were looking at doing some sort of reporting on the DIYs based on these identifiers. Now, the reporting never eventuated and there are other ways now to get reports on your DIYs. And there's also a new piece of functionality which I'll go into in a minute where DIYs can now be transferred between owners. So the client identifier sort of became redundant within the DIY streams. So we've decided to remove those out and it makes the DIY obviously a little bit simpler, a little bit clunky up, a little less clunky up, I should say. So the next change was around providing a separate prefix to each one of our DIY clients. So this was a request that came through from Datasite earlier this year where they wanted ANZ to allocate a separate prefix to each one of our clients. And they've taken this approach globally. So a lot of the other clients that are working through Datasite already have separate prefixes for their clients. And we wanted ones that were sort of lagging behind in updating our system. So Datasite asked us to do this for two reasons. One, it reduces the chance of identifying clashes when people are minting. So they're more likely to be unique when clients have their own prefix. And it will also make, again, the DIY transfer between owners a bit easier. All clients as part of R28 have been updated. So everyone has their new prefix. And DIY clients probably already noticed that these are sort of 10.2 something prefixes. And any DIYs that were minted with the previous prefixes, so the 4, 2, 5, 6 and 7 will still be able to be updated by everybody. So there shouldn't be any impacts to anybody in regards to this change. One thing I will just make a note of is the prefix that are assigned aren't necessarily, well, they aren't a client's prefix because they can obviously be transferred with this new functionality, which I'll discuss in a second. So they're not owned by a client, but they are allocated individually to clients. So the transfer of ownership. So some new functionalities being put in place within data site that allows or enables the transfer of DIYs between different clients or even different allocators. So ANT is obviously an allocator, but it could be that another allocator is signed up and we want to move to that allocator. So that can also be done. And it could be one DOI, it could be several DOIs or it could be all DOIs that a client has that need to be transferred. To sort of start the process, you drop an email to services at ANTS and they'll sort of walk you through the process. There's also some documentation up on the ANTS documentation space outlining sort of the steps that will be required. But one of the main things is there will be sort of an email or I guess a sign sort of acceptance of the transfer of the DOIs between parties. So the new party that's going to be the owner has to sort of agree to be taking these on board and be the manager going forward. The last one I'll just go on in the DOI changes. When we put out the proposal for R28, we also were proposing to move to a new method of testing through the DOI service. And this was the idea was to have a single test account for everybody. So there'd be just one account that people would hit. All the DOIs that people were testing would be listed under that single account. And we did receive some feedback from clients and some concerns about this approach. So we postponed that change and we're coming up with a new proposal that will be sent out in a few weeks time for comment. And then we were aiming to sort of release that in release 29. And we're sort of aiming for the lowest impact possible on our clients. So we'll be in touch. The next piece of functionality is the Research Data Australia Relationships Graph. So you can see on the right here this is kind of what it looks like and I'll go into the system in a second. So this originated from some work coming out of Research Data Alliance and it was a project which started with Research Data Switchboard. And this project basically aggregates information from different publishers and sort of areas of research output and draws connections between those to provide value. So we've sort of taken that idea into Research Data Australia for our own data and the graph provides the visual representation of the relationships within RDA. And one of the things that the graph sort of does that we can't see in the existing RDA was provide valuable information between the relationships between objects that can't easily be seen in a list and I'll show you some examples in a second once I get into the system. We're looking at a future enhancement where we'll actually be pulling information out of the Research Data Switchboard. So this will be things like where the Research Data Switchboard knows about organs that are connected to say research outputs like publications or data that we don't have in RDA so that will sort of be a nice value add. There are some known issues primarily with Internet Explorer at the moment. We're having some issues with the engineering of SVGs in IE, 11 end edge so we're working on that at the moment and we'll hopefully get fixed out soon. The graph does the display but it's not as pretty as it should be. So off to go over to the system so we can have a quick look. So here is a data set within Research Data Australia and I'll just scroll down to sort of where the relationships section starts within a data set record in RDA and you can now obviously see that we've got this part of the car which is the graph visualisation. By default when you come in it's kind of minimised down so that we didn't take up too much real estate within the record but it also uses a sort of indication that there is something there. You can hover your mouse over and it will tell you to sort of click to explore the graph but there's also this expanded collapse icon on the right here. So underneath is the sort of view that we've had in the existing system up until now so this is the list view of the relationships between the objects that are either within RDA or external to RDA. And the graph view basically is not a one-to-one representation of that list view because it does provide additional information but what we can see in the graph is, and I'll just reorganise this a little bit easier to see. We have this node here which is representing the current collection that I'm viewing or the data set that I'm viewing and it's what we call a primary node and it's indicated by this grey ring around the collection. You can hover over each of the nodes and it will give you the title and the type so we can see here that we've got a data set and this is the title for it so we can do that to each of the nodes. If you want to navigate off so if this object exists in RDA or it is a resolvable link so a link on the web the title of the node will be provided as a link and you can click on that and navigate off to say a website application or another object within research data Australia. The relationships between each of the nodes in the graph are displayed on this sort of connector so we can see here we have the association between two data sets and you can also hover over the connectors to find out the text for that relationship as well. There are instances where there are multiple relationships between two nodes so if I come over between this group and an activity I can see that the connector actually has a text multiple on it and if I hover over that I can see that there's multiple relationships in their express between the activity and the group there. One of the things that you can do within the graph so you can obviously zoom in and out within the graph you can also reposition the graph so when you start to get a really big graph you can zoom quite far out and reposition it to fit the whole lot in view. You can also explore other nodes with the graph so when the graph first loads we get the primary node and any sort of direct relationships to that node some other nodes which I'll discuss in a minute but there's also a way to explore these extra nodes that are displayed in the graph so we can see between this data set we have what's managed by this party which is John Scott. Now whoever's discovered this record they may have come from Google doing a search and they've discovered this data set and what they've found is that John Scott is actually the manager of this data set and I may like to know what other data sets John has connection to so what we can do is we can click on double click on John and this will basically attempt to load any other relationships that John has within rest of Starter Australia so this is brought in another node that's directly related to John. So from this you can see that this node is quite a bit bigger and this is what we call a cluster node and we basically put clusters within the graph when there are more than 20 single nodes with the same relationship to another node and they are the same type so if I hover over the cluster you can see here that we've got 103 data sets that are managed by John and one of the nice things in the clusters are that they tidy up the graph a bit so if I didn't have the cluster I would have 103 nodes on my graph making it look a bit like a hedgehog but it also provides a way to access these records within RDA so all these data sets they exist in RDA and the link up here will basically take me to a search result where I can then further filter those records down by all the other filters within RDA so it might be by subjects or by type that just makes a little bit easier to sort of I guess find what the user might be looking for. I'll just flip back to the graph in terms of highlighting the relationships within the graph you could hover your mouse over a node and it will basically gray out that anything that's not directly related to the node that you're hovering over so we can see this activity is related to almost everything except the data set on the right and the cluster above. One of the things that can happen is if there are no additional relationships and you double click a node it'll basically just sit there but this activity again the user might have come across this collection and they want to find out there's a bit of interest here because this collection or this data set is an output of an activity what else came out of this activity so I can double click on the activity like I did before for John and this will load any of the relationships to this activity. What we'll get is an extremely busy graph so I'll just zoom out a bit and now we can see this activity that we've just double clicked on has a relationship to a lot of other data and this is obviously a really good way to say there's more information within RDA than what you can see within the list view and if the user wants to drill down into some of these activities there's clusters in here or they can go to the activity record itself and then find out more about that activity in some of the outputs. Let's flip over to another example. This is just another example where we can see some other object types so I'm showing this example because both this object which is a website and this publication here that don't actually exist within RISC of South Australia but we can navigate off to those publications and websites based on the link within the tooltip. The other thing I'd like to mention is this was sort of the first iteration we are in the process of updating the graph already and that will come out in the next couple of weeks so there is I guess a limitation at the moment within the current version of the graph in that there is no controls to maneuver the zoom in and out on the graph so if the user doesn't have a scroll wheel they're really really stuck in terms of what they can do with the graph so this is sort of work in progress but you can see we've got some controls up here where we can zoom in and out of the graph and where we have nodes have lots and lots of connections but if we were to say we can resend the graph if it was quite big to fit things within the view. We also have a link to help which opens the research data of Australia help so there will be a new tab in the RDA help which sort of describes the different types of nodes that you can have and how to work with the graph and what's represented within the graph so that will be coming out in probably about two weeks time so keep an eye out for that. I haven't seen any questions in there. Jerry can you just let me know if there are any questions. I'm finding it really hard on this computer. We do have one on that particular aspect of the graph from Nick who asked is there a way to find out more about a relationship on the graph than just the label of the relationship like a definition for it. So I think that may have been answered a bit later in your presentation. Yeah, I mean there's not a lot in RDA in terms of the description of those relationships. We do obviously have pages on the ANS website and for the ROOF CS documentation which describes those relationships but that could be an enhancement we look at in the future to provide a bit more information within those pop-outs about the relationships. Okay so the next feature I'm going to go through is the ANS IGCN service so those who don't know what an IGCN is it's an international geosample number and they are globally unique system for physical samples and specimens. They're SOBL and they facilitate the linking between samples related data publications and people and the global sort of IGCN system is managed by the IGCN EV organization and technically it's underpinned by the handle system so that they work, they're very similar I guess as DOIs in that you have some metadata that's associated with the identifier that describes the object and they're unique, they're persistent and they're resolvable. So you can see underneath here I've got an example of how an IGCN looks so pretty similar to a DOI. We have sort of a prefix and then sort of a suffix which is actually the identifier portion I guess the unique portion and you can obviously resolve them spinning on a pin by the handle service you can resolve them via handle.net Let's go back to this again So in terms of the ANS IGCN service this was a collaboration between OSCOE ANS and CSIRO and it was implemented as part of the geosciences data and ANS virtual laboratory project It's I guess based upon the IGCN software that was written by CSIRO and enhanced a little bit for needs. It's free to use and intended for use by the Australian Earth Science Research Community It's accessed via the OSCOE website and requires AEF login at this stage Now when users go to IGCN they're required to provide, for when you're into IGCN you're required to provide a minimum set of metadata and also a landing page so there obviously needs to be somewhere where that IGCN resolves too When users min through the ANS IGCN service the landing page is actually created for the user based on the metadata that they're entering so there's no need for them to host the landing page somewhere on their own systems. It's all done within the ANS service If users are using the API service then they will need to host the page somewhere or we'll talk to ANS about possibly hosting that for them. We are obviously at the moment it is pretty much restricted or not restricted but intended for use by the geosites community but ANS is obviously interested in extending the service out for other physical sample types and I guess I should mention that the IGCN, the international IGCN is also looking at expanding things out for other sample types. So if there are anyone on the line that's interested in their own different domains then they can get a touch with ANS and we can talk through a process for extending the service The other thing to mention is that both CSIRO and Geoscience Australia also have IGCN services so the IGCN service is really focusing in on the research community, the academic research community which is not really covered by Geoscience Australia or CSIRO So I'll just go into the system for a second, hopefully this works. So as I said the way to access is through the OSCOPE website so I'm just looking into OSCOPE. This is the access point within the OSCOPE website It tells you a little bit about IGCN and some of the conditions that apply for using the service obviously has to be associated with Australian research need to provide a minimum set of metadata etc etc So to access the service there's a link down the bottom here which is clicking to access the ANS IGCN service and I'll just point out that there is also some documentation on the ANS website that's linked to from this page Click in, this will navigate me off to AEF where I can choose my institution So as I sort of mentioned before the service is free, anyone with an AEF can log in like I just did. You don't need to be affiliated with an organization like our other services at this point in time. So anyone that logs in through the OSCOPE website kind of gets an affiliation to OSCOPE So anyone in the Geoscience community we sort of track through the OSCOPE login Once you're in, you basically land on the page to mint an IGCN I'll point out just before we get into it, just at the top we have a link to some documentation So that's a little bit of help around the form You can log out, you can list your resources that you've already minted in the past where you can manage those so you can go in and edit those or view them and then you can obviously add another resource Now in terms of the form, there's a number of required fields which are highlighted in red and they're really around describing the core elements of a sample So the sample type the material type and who's the curator for that sample You can also set the visibility of the metadata So if I click on the drop down here, you can have the metadata for this IGCN publicly visible so that will obviously resolve all within the ANS service You can have it private so if someone does try and resolve it, it will actually get to a login page and they'll have to have access before they can view that And you can also put your IGCNs under an embargo for a certain time period and then once that embargo elapses the metadata automatically becomes publicly visible In terms of the other fields, if I scroll down a little bit, you can obviously describe location, pretty important for a lot of physical samples And then you can go in to describe related resources so you might have other contributors that you want to describe and you can put in some relationship types and contributor types between these objects Related resources so you might want to relate to publications and data or even other IGCNs, there might be a hierarchy between the IGCNs where you have a core sample that's then broken down into two individual samples that you sort of want to link together There is another information tab which is sort of access metadata around the IGCN so you can put in some specifications, the method that was used and also linked to a project and any other comments to do with the sample There are the link context help items for each of the fields, sometimes they make sense sometimes they're a little bit ambiguous but a lot of this is sort of the code that we took on board and we will enhance going forward Once you've minted one, so I'll just mint the sample and make it publicly visible I'll just fill out the base in here Once I've registered notice, it takes a little while obviously because we're hitting sort of the global system to register with IGCN and then obviously into the handle system You'll get a little pop out to say that your handle's been minted and you can find that the resolution URL here. You can view the metadata and another IGCN or just click OK which leaves you on this page and you can basically use the same metadata to register another IGCN with a few tweaks But I'll click on the view metadata and it will take me over to the view page and that's really big, I'll see if I can zoom out a little bit And this is basically the landing page that gets generated for your IGCNs within the work GUI So whenever somebody resolves one of your public IGCNs that they will land on, they obviously won't be logged into the system, won't be able to edit But this is the descriptive metadata that they will find and if I put locations in the map that would show up here Now if you are the owner of the IGCN and you were logged in, there is an edit button Or you can find in your list resources, you can find the IGCNs that you've minted in the past and view and edit those as you need So it might be that it was private and you now want to make it public Or vice versa if you edit the metadata and republish that IGCN So that's pretty much it on the IGCN This live we will be putting out a minor update which in the next few weeks where we are going to make it a little bit easier for users to link to orchid IDs and grant information related to samples So that should be coming out in the next few weeks So Joel we do have one question about the IGCN service Nick is asking is ANS posting IGCN metadata directly to the IGCN service? The reason he asked is only one handle server for IGCNs internationally and maybe there's a requirement for some redundancy I agree, no we are sending the metadata directly to IGCN and I believe that there is an application going in for some funding for IGCN, I don't know the exact specifics but if they get that funding I think there's going to be some work on the metadata kernel and making the system a little bit more robust but to answer Nick's question we are sending it directly to the IGCN service and they have a single handle prefix for that so there's a redundancy there So the next item is the RBA notifications so again this one was put into R28 and what I'll say before I get into the notification stuff we have done a huge amount of work for research vocabularies Australia in the last six to eight months, there is very little of it which is noticeable to the end user but basically the whole back end of the RBA was rewritten in the last eight months and some of the real benefits of that are things like the notification system that we can implement there's also an API which is currently being documented and will be released in the next few weeks so that will allow users to basically communicate with the RBA system machine-to-machine and publish vocabularies and it's also enabled us to put in lots of hooks for future enhancements that are sort of sitting in the backlog that people have asked for so we've done a huge amount of work in RBA so things are looking pretty good there. Now in terms of the notification system these allow users to subscribe basically to vocabs or vocab owners and it allows them to stay up to date with any of the changes or when new vocabularies are published So the list here users have the choice of subscribing to mailing lists for changes to a specific vocab or they can subscribe to any vocabs from that owner so that is new or changed vocabs or they can subscribe to everything in RBA so whenever a vocab is changed or a new vocabularies published they'll get a digest email to notify them. At the moment emails are sent out weekly on Wednesday mornings so when you subscribe you don't actually have a choice of frequency at this point in time they're just on a weekly basis on Wednesday mornings So I'll actually go over into RBA it just shows the list of the books basically I'll just jump into any vocab you can see at the top here in the title bar there's now a subscribe label so basically we can click that and it will open a subscription dialogue box and this is where you can basically subscribe and sort of add your preferences to the mailing lists so we can see here I'd like to subscribe to a weekly email digest and as I said there are those three choices at this stage where you can subscribe to the individual vocabulary. All the vocabularies from the owner so ANZ is the vocabulary owner for this agriboc or new or changed vocabularies from all owners within RBA. So there's a radio button group you select one of those. There's also an option here to be notified of any service updates and features so this is when we do service software updates to RBA you can tick that box and that will add you to the mailing list for RBA features and updates and that will be sent out as needed basis so whenever we do a software update to RBA you can send out an email to all the users on that mailing list it's pretty straightforward you just put in an email address it has to be valid obviously we do have capture in place most of the time the capture will realise that you are an end user and you won't actually have to do the capture clicking on images and then you can just click subscribe and you'll see this. I didn't need to do it and it will basically just say that you've successfully subscribed to that mailing list. So on a weekly basis let me just see if I can get this. There's always something that goes wrong. Two seconds. While Joel's sorting out that little technical glitch we have had a question from Jackie asking about whether some journals are actually requiring IDSN samples IDSNs for samples in order for articles to be published which is becoming increasingly a requirement for research data so as far as I'm aware that's not yet a requirement but there are a number of journals that will enable the linking from a journal article to an IDSN and they're actively supporting and encouraging that we do have a page on the AIDC website or the ANDS website that provides some background about IDSN generally, what they are, how they can be used how they can be cited and I'll pop that URL into the question pod so that people can follow up it may be useful too if you're talking to colleagues or researchers about IDSN just to give a bit more background on what they are and how they can be used. Sorry about that. No worries Joel over to you. Sorry I had the address completely wrong, I should have put it in a slide it would have been easier. So I've just come into the test mail server just to show an example of a weekly digest email that gets sent out so this is basically what users will receive if they've subscribed to whatever mailing list but no changes have happened within that week, they won't actually send emails so it's only when things actually change that impact the preferences they have. So here's an example of a weekly digest so basically the email's sort of broken into a number of sections, I don't know if I've got one here that's got, sorry to be fiddling around so this one, you can see there's a section here which is to do with vocabularies that you're subscribed to directly, you can see here the name of the vocab and there is a link off to that vocab directly within RVA and then we can see a list of sort of the changes that have taken place over the last week so it will tell you if the status of the vocabularies changed, it's not obviously here and then it will tell you the descriptive metadata elements that were changed so we can see that the title was changed within the last week and then it's broken down a little bit further into the version changes so it'll detail sort of the things that have changed within each of the versions so we can see that a new version, another version here was added and it's set to current which means that the other two versions that were there would have been updated to superseded. There are some changes to NEI version one here the version is no longer published by the link data API so some of the flags within the CMS in RVA were changed for this version and then down the bottom of the email there is a button which allows you to obviously change your subscription preferences so because there's no requirement for users to log in when they subscribe the only way that they can sort of get to manage their subscriptions is through their email so until you get an email you can't actually update your subscription preferences so you have to wait for the first email to come through. The other example I've shown before the other section that's within the email owner changes so where you've subscribed to particular owners or all owners within RVA it'll break down the vocabularies under each of those owners so I think this one and might have several so here's answers the owner and we have the different vocabularies under the owner that have changed and we can see here this one actually had quite a lot of changes within this last reporting period. I'll scroll to the bottom. The subscriptions button and that will take you back to RVA and this is the managed subscriptions page where you can obviously unsubscribe from each of your preferred mailing lists. Again it's broken down into that sort of two categories so we have the vocabularies that you directly subscribe to and also the publishers that you subscribe to. Now I'll just point out that the two options down the bottom here we have all so this is that sort of special category where you want to be notified of everything but you can also specify that you're interested in just vocabs from this owner so the second or third option down here the AU testing role is not really required at this stage because you're going to get everything anyway from the first option but we can obviously unsubscribe from that one and leave AU testing there. It's pretty much as simple as that. Where you are subscribed to the service updates so the software sort of updates notification zero is also another option that's displayed on this page which enables you to unsubscribe from that mailing list as well. So I think that's pretty much it. There is obviously some documentation around each of these pieces of functionality so I think we're going to send out the slides so the links will be in there if you want to sort of investigate a little bit further and read some of the documentation. But that's pretty much everything. Are there any questions that didn't answer or have? Yeah thank you Joel and thank you for bravely doing so many live demos. It does enhance being able to see the actual functionality. We do have one question answered and it's back on the IGSN service from Liz who asks can the metadata be extracted from an IGSN over some kind of service? Yep absolutely so one of the requirements of running an IGSN service is to have publicly accessible OAI BMH endpoint so we have that running. Any of the metadata associated with public IGSNs or ones that have embargo periods that have expired will be available via that service. That service will also be used to feed into the Australian IGSN portal which is run by CSIRO at this stage so they'll be harvesting from us as well and that will be discoverable through that service. Well we don't have any more questions come in now so I think we'll give ourselves an early minute. Thank you again Joel for your time and for running us through what's been a lot of work being done by the dev team and thanks to all our participants for coming along today for your questions and we certainly hope to see you at another webinar again soon.