 Good afternoon, everyone, and welcome to today's webinar. My name's Jerry Ryder, and it's my pleasure today to introduce you to our presenters. We have four speakers. Jens Klump will be our first speaker. He is a geochemist by training and is science leader of Earth Science Informatics in the Mineral Resources Unit of CSIRO. And importantly, he's also vice president of the International Geosample Number Implementation Organization, which you'll touch on a little bit in his presentation. After Jens, we'll hear from Brent McGinnis, who's a research professor at Curtin University and director of the John Delay Center in Perth. He has more than 26 years of international experience in the geoscience and resources research sector. Then we'll hear from Joel Ben, who's a manager of DevOps with the ARDC, where he's worked for more than five years as part of the development team delivering online services such as research data Australia and the DOI and handle minting services. And Joel will be giving us a live demo of the IGSN system. And then finally, we'll hear from Leslie Wyborn, who's based at the ANU with the NCI and the Research School of Earth Sciences. And Leslie will be talking to us about the future of IGSN. So without any further ado, I'd like to hand over to Jens as our first speaker. Thanks, Jens. Thank you, Jerry, for the introduction. And thank you for inviting me to speak today. I would like to give you a short introduction about what IGSNs are, where they come from, and what you can use them for. Just as a very dry definition, IGSNs are persistent, globally unique, web-resolvable identifiers for physical samples. Some people call physical sample specimens. Anyway, the pictures on the slide show you some applications where IGSNs have been used. It's quite a wide range of samples. They're not all geological. They can be used for anything that is a physical sample. And that is an important aspect for the future development. So what does persistent mean in this long string of this definition? Those who have been around on the internet for a while have noticed that URLs go away. Sometimes we need to change the arrangements of servers and how they're organized, and whatever technology changes. Things change. And over time, that links to a phenomenon called link rot, where links just don't work anymore. And persistent identifiers were designed to deal with this problem. So what they do is they will always resolve to a so-called landing page, even when the URL changes, or sometime in the future, when URLs will no longer be used. HTTP as a protocol on the web has been around for 25 years. And I don't expect it to be around for another 25 years. But the record of science, we hope, will be around for centuries. So there are other examples of persistent identifiers. And the most popular ones you will probably know is DOIs, digital object identifiers, and handles. They're actually technologically all the same, but I won't go into that today. A very important aspect is this aspect of identifiers being globally unique. And this map gives you two examples of what uniqueness might mean. On the left-hand side is a list of identifiers that can be found in the literature for a sample that has this strange name of Argon 003 in the PetDB database. And in the literature it has about a dozen different names where one can assume that they all refer to the same object, but you can't really be sure. So having an identifier that gives a unique identity to the specimen would be really helpful. The example on the right-hand side is the map of the world listing all the samples from Earthchem with the name M1. So M1 seems to be everywhere, and especially in Japan. And as a material it's anything geological. So M1 obviously isn't a globally unique name, and it's also not very useful to identify materials. So there's this aspect of web resolvable. And just at the top of the page I gave you the URL of the page where I took the screenshot, something long complicated that is likely to change in the near future. But just below that is the IGSN, the arrow points to the big bold IGSN. And underneath is just as an example how this particular IGSN is resolved by using a handle net resolving server. Usually this should all be encapsulated in the applications and you shouldn't worry about it, but sometimes you need to. And this is an example of how you would resolve it. You can actually do this with any DOI or handle through handle net or DOI resolvers. And then at the bottom of the page is an example of what this specimen was called in the field. So somebody in this Hammersley Basin project called it HM-96B, which might or not be unique. But what I want to show with this slide is that you can call things in your applications in the lab or in the field with a locally unique name, something that makes sense to you. But to make it globally unique, this is where the IGSN comes in, like maybe a passport number or social security number for a specimen. So why would we want to use IGSNs? As I already mentioned, sample names are locally unique, hopefully. That should be part of your good housekeeping. But they're usually not unique in a global context. And so what it enables us, once we assign these globally unique identifiers, we can use these identifiers to make exact references to samples and data and literature. So the thing that I showed you from the Pacific with a dozen different names, if we had an exact reference, we would know what we're talking about or we would be able to refer the many M1s to specifically individual samples. And having all these samples uniquely identified would also allow us to compile large scale catalogs and overviews for overview studies. Very practically, having identifiers would also help us track samples through varying institution samples. Sometimes do move around. They move between institutions. They move from a lab to a repository. They move from one lab to another. And so this can also serve as a tracking number. It's not just another label on a box. And so because we can now tie a sample to a data set to a publication, it also allows us to verify sample-based data. Many models use some data set based on some sample as a ground truth element. And it's at the moment very difficult to go back to that ground truth. And having an IGSN would help us finding those ground truth evidence bits. So what can I use the IGSN for? As I said in my opening statement, you can use it for any kind of physical sample. Historically it's the international geosample number but it's not limited to geological samples. And so we're thinking of how we can re-engineer the acronym to mean something more global. But you can also use it for something that is not physical. Something that we call a sampling feature that is the thing where the sample is taken from. So a borehole is not a thing. It's a hole in the ground. It's not a physical object, but it's a sampling feature that we need to identify to relate the cause that were taken from this hole to this borehole. Or a site which might be an outcrop or a sampling site somewhere where something was taken from nature. This might sound like a bit of a distraction from physical objects, but we need this to identify where things came from. And since there's nothing else in place, this was a very pragmatic decision to also make this a feature of the system. Same, also not a physical object is when you aggregate samples into some kind of a collection, like a box of samples or a string of pieces from a drill core or some things that work hard together in the net that have some, share some association. This aggregation of samples can also be identified with an IGSN. And then very common procedures that we take sub-samples from an existing sample. And these can also be identified and then related to the parent object. So IGSN can also link to other IGSNs like the parent-child relationship or specimen to collection or specimen to feature. We can also relate them to DOIs in data and literature and to other persistent identifiers through element called related identifier. And this is a feature that we share with data side. So we can, data side objects can also refer to IGSNs. So there's this mutual connection between the two systems. And this figure I want to illustrate to you how things can then relate to each other. So we have a specimen that's identified with an IGSN. On this particular piece of kaolinite, we measured an infrared spectrum. This data set is then identified with the DOI and stored in the CSRO data access portal. And then this spectrum is part of a publication where it is interpreted. This publication is identified with the DOI. And then you can go back from this publication with this provenance trail and go back to the specimen and find out what was it that was analyzed. And in this particular example of spectra measured on rock specimens, we had a project where we made sure that the thing we measured was actually what we thought we were measuring. So that kaolinite was actually kaolinite and not something else. And if you had any doubts from the publication, you could go all the way back and find the sample where this data set was derived. To give you a bit of background and global context on IGSN, IGSN and data site evolved about at the same time in the mid-2000s. And when we had some conversations about how to implement this around 2007, samples were seen out of scope for the precursor of data site. Data site was founded in 2009. So this is the historical reason why this is not one system. But there are also other reasons at the moment still in the way we govern the system that make it better to go in parallel but not as the same organization. The IGSN implementation organization was founded in 2011 and at present has four members in Australia, CSRO, who joined in 2013, Geoscience Australia who joined in 2014, Curtin University joining in 2015 and Australia Research Data Commons back then still ends in 2017. So that's a bit of an illustration. There are about 20 members in a number of countries on various continents starting to show kind of a global spread and it's showing some impact that Leslie will be talking about at the end of the webinar. There are other persistent identifiers around. I've already mentioned data site but in relation to identifying specimens, I think IGSN plays a really important role because at the moment we have 6.5 million samples registered. There are a couple of hundred registered with data site but it's important to point out that data site metadata are not suited for samples. They are very well suited for bibliographic objects and data but not for physical specimens. A very big player in this area was the Life Science Identifier, LSID, but it was kind of ahead of its time and that limited adoption and eventually the system broke down and was discontinued last year and the biodiversity community reverted back to a URL based system which they say is just an intermediate measure until they find something else that works but they have to find something now so they step back to using what they call cool your eyes. And that's all I wanted to tell you about IGSN from my perspective and now I hand over to Brent to give you the university perspective. Good afternoon everybody. My name is Brent McKinnis. I'm the professor of economic geology and geochemistry and the executive director of the John Glater Center Lab at Curtin University. I also have a role as a national coordinator of Oscopes, Earth Composition and Evolution Program which involves multiple laboratories around the nation. I'm speaking to you today with two hats on. The key thing about the John Glater Center, if you don't know, it's a centralized analytical facility at Curtin University. It consists of 14 laboratories, 25 staff and they're operating over $33 million of high-end analytical facilities. These labs provide analytical services to over 550 unique researchers. The labs operate for these provide research services to these individuals about 27,000 hours a year of analytical time. So the data volumes being generated are quite high. The volumes could not be considered high relative to perhaps satellite data or something. Probably we generate about 10 to 12 terabytes of data per year which is actually not very much but the data is high value data and that the data can be used to assess perhaps the petroleum potential of a basin could be able to provide you with the age or thermal history of a rock sample or mineral belt or it could provide information perhaps about the health of an ecosystem. So with that perspective, we have a problem in terms of data management and data delivery and what we realized that we needed to do was we needed to come up with a better system of disseminating the data sets to our researchers than simply allowing them to take the data on a thumb drive or a hard drive. And so we wondered if we could create a system in the laboratory that would automatically capture data about the samples and make it available and reusable in the future. So as an example of the instrument suites that we have at Curtin, a lot of them are mass spectrometry systems or imaging systems such as scanning electron microscopes or TEMs. And so these are the sort of instruments we're talking about in the operational hours. And in general, each researcher has between 10 and 100 samples per year. So you can imagine there's a lot of information if you're trying to capture it in a systematic way. So we did a trial. Here's an example of some samples that were collected from the Geological Survey of Western Australia, the samples on the right hand side. Think of them as grains of sand. They are cast in epoxy mounts. In total, there are 5,700,000 grains of sand in those samples collectively. But what they are is they're heavy mineral concentrates that the Geological Survey collected when they went into the field to do their mapping and survey programs. And we were able to use this instrument in the bottom left-hand corner to automatically measure the mineralogy and the chemistry of those samples after eight days of continuous operation. So with 200 samples here and a lot of individual analysis, we were automatically faced with a sample and data management problem. So we worked with Oscorp and ANS to come up with a system of figuring out how to actually identify the samples because we saw that there was going to be a problem with the M1 problem that Jens talked about if we started systematically doing these analysis for other projects that there would be, theoretically, a high probability of duplicating the sample numbers. And quickly realized that the IGSN system was the best system to adopt for the project. So the number of things that we've done is we've incorporated the IGSN number into a QR code system on the sample which permanently identifies the sample. We've actually made that data set available so you can actually go and look at the data set and explore the mineralogy. The minerals are classified by chemistry. So on the far right-hand bottom corner of that slide, you can see that there's different colors and those different colors represent the chemistry which is then reflected in the mineralogy. So anything red is a titanium dioxide or a rutile, a grain, anything blue is a zirconium silicate, and so forth. You can actually go to the website and find information about that. One of the things that we found out is that over 15% of these samples contain minerals of economic interest. And so that could be of interest to exploration companies or certainly people are now wanting to access these samples to do enhanced geochemical studies on them, looking at neodymium isotopes and some strontium isotope data on some of these samples. So now for samples that were previously stored on the shelf at the Geological Survey of Western Australia, they're now available for reuse and interrogation of the data sets and not only mineralogy but you can get chemistry data. The IGSN number is for this particular sample here is actually embedded into the QR code. So you can actually take your smartphone and zoom in on that QR code and all the sample information comes up along with the data analysis. So we like to think that this is something that can be used not only at Curtin but used at laboratories around the country. So from a research perspective, I wanted to give you my personal opinions on how this will change if not revolutionize the way we collaborate and cooperate. I've put on here a board of slide from EA, Elena Bastercova. And this shows where you can the IGSN implementation at Geoscience Australia where they have over 1.6 million samples registered. The fact that they do this makes the samples discoverable online. And the perspective here is that if we were all to do this, we would actually be able to know where can I get samples or who can I collaborate with? Is my idea original? Is somebody else already done work in this area? Do I need to add into my proposal a fieldwork component if I can access the sample somewhere else? In some places in Australia, it's very difficult to get access. Permitting issues can take time, a lot of bureaucracy. And so if you can actually, if the samples exist somewhere else, it's good to know. And I think it will benefit people who are interested in true collaboration. Maybe you're an expert in one area of mineralogy, but you're not an expert in isotope geochemistry. Somebody might contact you and say, hey, can I access your samples and collaborate? And I think it's going to really change the way we do business. And the other thing is that when you review grant proposals, you might look at whether or not they actually need to get fieldwork funded if those samples might exist somewhere else. So I think that in the long term, IGSN will actually change the way we do business, how we work with each other, how we win grants, and how we receive recognition for our work. So I'm really excited about it, and I hope you are too. Thank you. It's Joel Ben from ARDC. I'm going to give a quick overview of the ARDC IGSN service. So the IGSN service was implemented as a collaboration between Oscopans and CSIRO, and it was implemented as part of a Geosciences Data Enhanced Virtual Laboratory project, and that project was funded by ANS, Nectar, and RDS, who have now formed Become ARDC. The service itself is free, and it's intended for use by the Australian Earth Science Research Community. Now, it's a service that is, I guess, self-service. So your institution or organization doesn't have to maintain the account or sign up and manage software in any way. It's hosted by ARDC, and anyone with an AEF login can access the service and win my GCNs for their samples. It's accessed via the Oscopan website, and as I said, it's accessed via an AEF login. We do have means of providing users without AEF logins access to the service as well. So to do that, you'd need to get in touch with the ARDC. In terms of minting an IGSN, there's minimum metadata that has to be provided in terms of minting, and I'll go through that once I get into the actual web form in a second. And I also just wanted to point out, being a persistent identifier, there's also some responsibilities for the, I guess, the minter of that IGSN to maintain the metadata within that IGSN. That's the descriptive metadata of that IGSN. And also the ability for that IGSN to resolve. Now, the resolvability side of things, when you're using the ARDC IGSN web interface is taken care of for you. So when you mint an IGSN, it actually creates a metadata page for you, and that is where the IGSN will resolve to. When you're using the API, it's a little bit different. You will either need to host the page that gets resolved to, or we can do something for you in a future release. At the moment, the service is really obviously, because it was implemented as part of the Geosciences project, it is really limited a little bit to the Geosciences samples. And that's mainly because of the metadata sort of schema that's being used to describe samples is for Geosciences. But we are very interested in extending the service out for the domains and disciplines. And the best way to sort of get in touch with us would be via the ARDC service desk. So services at ans.all.au. And we can sort of talk through that process. The last point is just that we do have an API available from machine to machine minting. So if there are services or tools that are creating lots of samples or want to do this in an automated way, but there is an API available. And again, getting contact with services at Ans, and we can set up accounts and go through how that would be tested. So what I'm going to do is just go through how users access the service. And as I said, it's via the Oscope website. So I'll just flip over to the Oscope website. Please let me know if they can't see the browser. So this is the page on the Oscope website. So you can see the URL at the top there. It's Oscope.org.au, idcn-info. And this just gives you a little bit of a sort of info about the service and sort of the, I guess the responsibilities of the users of the service. And then at the bottom, there is a link to access the service. So when I click that, it redirects me off to Rapid Connect or AAF Login. And that's where you can pick your institution to log in. So scroll down to the ANU and then click Continue. And then this will bounce me to my organization's AAF Login, where I connect to my details. Password, right? Once I've logged in, I'll land into the service. So that's, I've accessed via the ANU AAF and it's basically once I've authenticated, it's redirected me into the ALSC IDCN service. So that was the login process. But what I'm gonna do is just flick over to the test system so that I'm not actually minting production IDCNs as I go through this. The form that you land on when you log in is sort of the registration form. So this is the form where you will fill out the metadata to mint an IDCN. And as I said earlier, there are a number of sort of minimum fields that you have to provide in order to mint an IDCN. So you have to provide the object type. So there's a set of options and it's defaulted to physical sample. And then you can see the fields that are highlighted in read and these are the mandatory fields that you need to provide as a minimum. The metadata visibility, so you have a couple of options there. You can have it publicly visible, completely private. So no one will be able to find that in terms of the ARDC IDCN service and also it won't be harvested out through our OAI endpoint publicly. There's also the option to put it under an embargo period where you can enter sort of the end date for that embargo and when that elapses, that metadata will become public in the service and also via our OAI endpoint. You have to specify obviously the sample or item type and also the material type. So there's a drop down, there's obviously a number of sort of types there to choose from which come from known vocabs. The other thing to point out is there obviously is some in-context help. So there's little icons with each of the fields where you can click and it will give you a little bit of help for filling out those fields. The other field that is required is obviously the curator. So this can be an individual. It could be say a role at an institution. And it's obviously good to have that information up to date. If someone does find the metadata for your sample, they can get in contact with whoever it is to get access to that sample or find more information about that sample. So this is really important to keep, as I said earlier, up to date information in. Now as we go further down, there is additional metadata around the IGCEN. So obviously location is quite important for a lot of samples. So you can describe the location of where the sample was collected or taken. Related resources, so this is I think Ian sort of touched on this earlier where there's the ability to link between resources. So this might be a link between say two DOIs or two IGCENs themselves. So there might be a hierarchy between those samples and you can do that in the related resources. And the contributors, you can obviously put in anybody that's contributed in any way to this sample. And there is a number of roles that you can choose from. So there's things from Funda, Irina, Stakeholders and you can describe those by putting in their name at the role and there's even a spot there to put say their identifier, which could be an ORCID or an NLA identifier. On the last tab, there's other information to I guess extend the richness of the metadata around the IGCEN. So alternative identifiers. So this might be, you have a local identifier within your systems to represent that sample and you can describe that here. Classifications, purpose, the sample feature, so the feature that was sampled to get this actual sample or core or whatever it is, the data collection you can put in here as well. And also link to the method of collection, the project. So this is where you might want to link to say a grant that or a project that is basically funded the collection of the sample and any comments you can provide in here. Now going back up to sort of the minimum metadata, it's obviously to get your IGCEN, that's all you need, but in terms of having a really rich metadata record for discoverability and I guess assessment from other users, it's really important I guess to fill out those other fields further down the page. Now once you have minted a resource, you have access to all the resources under this list resources page. And you can see here that I've got all the samples that I've minted previously. I can go into any of these and I can edit them. So this is what I've done before. All the fields are filled out. You can see that I've got spatial location and you can basically change any of the metadata in there. So it might be that the curator has changed or the institution or there's additional information in terms of the project that you might want to add. And I'll just click update to take me off to the metadata view of this page. So this is basically how I'm updating. It's the same sort of process as you go through to mint. You'll get this pop up where it says that it's been minted. In this case it was actually updated and you get a couple of options. So you can view the metadata and that will navigate you to the metadata page for this IGSM. You can add another that will give you a blank form to add another one where you can click okay. And basically okay leaves the form filled out with all the same information because you may have another sample that's very similar and you can basically just tweak the information and mint another IGSM with that option. In terms of the IGSM resolve, being able to resolve the IGSM sorry, it's underpinned by the handle service. So you can use the handle infrastructure to resolve the IGSMs. And you can see here that we have the full resolvable link for the IGSM that I've just updated. And if I click that, it'll basically take me to the same pages I would have gone to with that view metadata link. So this is where I said earlier that in terms of using the ARDC IGSM web interface, we provide that landing page for your IGCENS for you. So you don't have to worry about hosting a page anywhere on your systems. So this is the view page for the metadata. So this is what someone will resolve the IGCEN to or will discover through a service. And it's basically all the fields that you've just filled out are available in a read-only mode. If you are logged in, there is an edit button so you can go in and make changes. If it's not publicly visible, so if you've set it to private or it's still under embargo, the user will actually, when they resolve it, they'll come into a basically a login page. And if they don't have access, they won't be able to view your metadata. There is the QR code up the top here. So if you did scan that with your mobile device, it will come up with the full resolvable URL for the IGCEN and then you can obviously access the metadata. I think that's most, there's a link up the top obviously to documentation. So there's help on how to fill out form and how things work. There's also some information on the API in there as well. I think the only other thing to point out was when you mention IGCEN, you actually receive an email to say that you've minted and a link back to the system so that you can sort of track and edit those going forward. And the email that's used is the email that's associated with your AAF login. So that's where you'll receive those emails. Well, I think that's pretty much everything for me. Thanks Joel. I'll just hand over now to Leslie. Okay. So I just, as Hien said, that Kirsten and I have been kind of developing this system and playing around with it as a heap of other people. It was about 2007. And we designed it originally to be within the Geoscience community. We knew there were life science identifiers and other systems around. And we were happily polluting around until all of a sudden we realized we were almost the last man standing. And what we also noticed was that people were starting to use it for all sorts of things just because they're desperate to get an ID system for a physical sample. And I wonder whether it is the ability for IGCEN to link to samples and data and drive from the sample into publications has made it extremely popular. And it is now being used beyond the original intention for Geoscience domains. And the growth is phenomenal. As Hien said, it's up to 6.8 million samples. We've had inquiries from museums, from anthropologies, et cetera. And it was getting too big. And the growth into the other domains needed to be moderated because you can't just take some of the Geoscience standards we were using them and applying them to bio and other samples. And so what we thought was, can we identify a key common kernel attribute across the domains? We also knew that the governance model needed to be strengthened because it was bad enough within the Geoscience domain but it was spreading to other domains. And we were starting to have not very good what I would call community behavior. People were actually minting doys but not registering, so minting IGSNs but not registering them. They were not following the agreed protocols. And then independently we had a few examples of people say developing vocabularies and other technical things and calling them IGSN. And at no stage had they put them through the IGSN technical community or let the governance body know that this was happening. And people were starting to put their own spin on things so we didn't really have a consistent message about IGSN and what it's for. So Santa Claus comes around, i.e. the Sloan Foundation. They heard us talking about this at one of our outreach events and they asked us to consider a grant. And we heard about two, three weeks ago that we have been given this grant for 20 months starting on 1st of August. It's going to establish a solid executable plan for the future of IGSN that will enable new organizations to participate easily and confidently but more importantly our system will work in that physical samples will be integrated into the research data ecosystem. So you can see on the right one of our key things is to identify what is that common kernel. And then from there you can work out as Jen, as Joel said, we've got these fields for extra descriptions, make the metadata richer, which we've done hopefully through a series of controlled vocabularies that are specific to the domain that is setting up the sample. And on the right you can see I've got the common kernel and I know for example with marine samples, something like grain size. So they can borrow the grain size vocabulary that sits in the geological GESIML vocabularies. And there's a lot of flexibilities but what we want is that key common identifier, that key common kernel that'll go across multiple domains. And so what we're going to use the funding for is a series of workshops for international experts. And data sites are still with us and are coming on board. So again, we'll keep that parallelism that we have had with data site to make sure what we come up with, we can integrate with data site. And we're going to slightly redesign it and its management to be able to use IGSN with confidence. And the experts are from the US, the UK, France, Australia. The representatives from Australia, from CSIRO and ARDC. Japan and South Africa. And I've given you a link to a blog which is kind of describing where we're gonna be going in the next couple of years. And we think it's exciting. For example, in bio, we now have TADWIG on board. And archaeology is coming out of Macquarie University. So I think it's the best thing that's happened to us. And yeah, I think I'll just finish there and hand over it back to Jerry. Whoever I do that. Yeah, thanks, Liz. If you just move on to the next slide, that'll be fine. That's just got some links on there that people can follow to find out more. So thank you to all our presenters for a really informative set of presentations. We've now got time for questions. We do have one question already from Barbara. And I think it may have been for you, Brent, where she asks about whether you put the human readable IGSN on the sample, as well as the QR code. Would you be able to respond to that one, please, Brent? Yes, both are there. In fact, we have three ways of identifying the sample. There's the original sample identification number, which you may assign in the field. It could be your own personal number. Then there's the IGSN assigned or minted number. And then where we can, we'll print out a QR code and attach it to the sample as well. Thank you. Now, this one may be for you, Jens. Andrew is asking how are the IGSN numbers assigned? The examples used don't seem to be random. That's a very good question. And that's a core question in the way we set up IGSN. One analogy that I like to use in how we make sure and how we allocate the names is the way that license plates for cars are allocated. So there's a system in place that makes sure that license plates in Australia are unique for every car. But how this is handled is very different from state to state and territory to territory. And this pattern that's called hierarchical delegation, this pattern allows us to accommodate the specific needs in the specific applications of IGSN. So one example here for field campaigns, we have little booklets printed like raffle tickets with unique numbers. And then when the sample is taken in the field, this raffle ticket is put into the bag with sample. In other campaigns, we've pre-printed labels with IGSNs and QR codes that, and then the QR code could be scanned into a mobile app that helps capture the metadata for the sampling process. In other cases, the samples already exist with locally unique identifiers in the database. And then we use a prefix before the locally unique database number to make this a globally unique identifier. So the way this is handled depends quite a lot on the specific situation and the governance that we set up is exactly tailored to be able to handle all this variation and how we identify samples. Thank you, Jens. Another question, maybe for you again, from Anthony, who owns the metadata and how is intellectual property being dealt with? So there's two ways to look at this. The one is the merit. Of course, whoever creates information has an ownership in it. From a legal perspective, as in other catalogs, the metadata are open to everybody because otherwise you can't compile them into global catalogs and you run through all sorts of legal entanglements and licensing problems. If you put restrictive rules on the metadata. So I hope that answers your question. From a legal licensing perspective, metadata are open, but of course we acknowledge the intellectual merit that goes into creating them. Thank you, Jens. It's always a tricky question about intellectual property. Joel, a couple for you. Helen wants to know, is it possible to upload a large group of details that are similar but need IGSN, for example, from a spreadsheet? So where there's repeating data, but many samples? Not at the moment. So the bulk functionality is only really available if you wrote, say, a script with the API, but it is something that we have without other services, so we would be looking at that sometime in the future, no doubt. But at the moment, the only real way to do that sort of bulk in what would be by the API. Okay, thanks. And I know your team, Joel, is happy to help people get the API up and running. Yeah, of course. And Barbara's asking the best website to go to for the metadata scheme and associated vocabularies. You can go through the documentation page, which is linked from the service or through the ANS website. There is information on the IGSN service. We find the metadata scheme. The metadata was, the scheme was developed by Siro for their project needs, but it has been accepted, I guess, by the GO community so far. I think there is work going forward to review that and the vocabs that are used. And obviously, Leslie also touched on some work on that sort of call metadata kernel that will happen. But yeah, to answer the question shortly, the ANS website would be the best place to check. Okay, someone's expressing a concern with the different levels for which an IGSN can be used. For example, borehole sample and subsample, how is that resolved in detail? That's not a clear question to me, I'm afraid. Jens. So this example of boreholes and materials coming from that and then being subsampled, this is very nicely illustrated by a use case from the International Continental Scientific Ruling Program where they've used IGSN to identify boreholes in one of their projects and then all the materials that came from these operations. And the link between all these objects is the related identifier element in the metadata where you can refer from a parent object to a child object or from a child object to a parent object. And in the IGSN documentation, on GitHub, this example is listed and when you resolve this IGSN, it takes you to a very nice page that I had on one of my screens, which actually has this tree structure of dependencies between objects, very well illustrated and interactive. So you can go from one level to the next and from one object to the other and display the associated metadata in pretty great detail. Okay, and a follow-up question. I'm not quite sure who on the panel can answer this one, but the demonstrated way to register an IGSN would take at least a minute. If GA had one person doing nothing else, they would be able to register about 100,000 samples a year. So how did they manage to register 1.6 million? I think that was actually globally rather than GA. Well, GA did register more than 2 million samples and they did that through using the API. The API was actually the original way to register IGSN and adding web forms was a later addition to accommodate the needs of individual researchers. Our original, not original, but our initial target group to grow the system quickly were large organizations like national geological surveys with large collections and they would then use the API for sample registration. Okay, thank you, Jens. A comment that it seems that the effectiveness of the services strongly dependent on institutions maintaining a museum or archive system for their samples, which is unlikely in many institutions because of costs. Does anybody on the panel wish to respond to that? Brenton, probably me. One short remark is that when we talk about persistence, there's sometimes a confusion between what is supposed to be persistent and whatnot and what we require in the system to be persistent is the description. The sample itself might not be available. Sometimes we do destructive analysis and after that analysis, the sample is gone. So you won't be able to keep it but you want to keep the information about it and its identity and that minimum is something that is what is required. But there are other aspects to it that Brent and Leslie might also refer to. I guess what I could say is that, yeah, this is a phenomenal problem around the world and in geology samples are being ditched. They're very expensive to store. And so one of the turning it around is, as Brent said, maybe we can cut down on the number of samples if we could carefully curate what they are, like how many trips do you need to the Galapagos Islands to collect those samples or if they're properly registered, they can be reused. And so we're starting to build a case for maybe fewer samples but better managed. And the other more important thing, as Jan said, is now that we can share samples so easily through this system, like they're not boxes sitting in people's cupboards. We are bringing to the front the fact that samples, if they're properly curated, can be reused and it makes it easy for you to go to the funders to say, well, here's a suite of samples that should be preserved because they're well described, they've got identifiers, we can connect to the data. And the funders can actually start to see another purpose other than boxes and boxes of rocks that have got the numbers M1, M2, M3. Also, when you take the national collections, like the national insect collection or the national herbarium, these collections already exist and are reference collections. And I see a lot of value in identifying these samples. The biodiversity world has been working on this for many, many years and unfortunately their attempt at the Life Science Identifier broke down last year. I won't go into the details of why this happened, but we're making a new attempt together with the biodiversity community to make sharing of these resources easier. Thanks, Jens. One last question before we, oh, sorry, Bryn. I just like to add my personal perspective. And I think we can all share horror stories around samples being chucked out. And so we've started a pilot project where we identified researchers who are near retirement, who have substantive collections at institutions. And we do recognize that those institutions may not have the resources to host those samples indefinitely. And so we're brokering arrangements with organizations such as geological surveys to actually store those samples or host those samples and to make them available. So I think in the future, we're gonna see much more of a consolidated effort to preserving samples which are precious. And if they're not registered in IGSN, no one will know that they exist. So I think it's one of those chicken and egg things that we just need to start registering samples. If you think your samples are important and you'd like to share them, then make them recognizable. Thank you, Brent. And that actually takes us to the end of our time. And unfortunately, we do have some unanswered questions. So we'll attempt to respond to those offline, to the individuals who have asked them. Before closing, I'd just like to say a big thank you to all our presenters today and to our audience for their participation. The fact that we didn't get through all the questions suggests that there is a lot of interest in this topic. And please, I would encourage you to have a look at the information about the service and about IGSN that it is available on the ANS website and to contact us if you have any questions about using the service or about IGSN. We hope to see you at a future ARDC event. We do have quite a series of webinars and events that you can find on the ANS website. Thank you again to our presenters and have a great afternoon.