 Let's hope it should be out there we are. Okay. Okay. Wonderful. Thanks very much Steve. Good afternoon everyone. I'm really quite honored and excited to be part of the conversations this afternoon. And I'm hoping that I might ease everyone back into the post post lunch break with probably something that's probably a little bit less technical than where most of you are. I wouldn't call myself a data scientist at all. But I sit in a really interesting space across a number of organizations involved in place naming. So I had a little bit of difficulty with that. With that mentor I must say Steve and I probably quite a few tick boxes in there on my part and I'll explain a little bit of that context in a moment. I think as it says up here I'm really I'm here today to just share some examples around place names and vocabulary use across Australia and some of the work that's happening in that space. But and this is going to come from I guess my perspective across having roles in a number of different organizations and they all have a bit of a bearing on the place names and vocabulary sort of overlap there. So before I jump into specifics for Australia though I wanted to provide a little bit of that context. I'm actually going to jump back a little bit into some some international context just to help you all understand a little bit not just where I'm coming from this afternoon. But also I guess the organizations and people involved in place naming in the vocab as well. So at the global level we start with a UN group of experts on geographical names and I thought the easiest way for me to explain what that group is about is to show you their vision which is that rather rather big mouthful of words that you've got on the screen there. Most probably relevant to the conversations this afternoon is the second half that I've put involved and it's really about having place names easily accessible for national and international use which of course this relates to consistent worldwide use of geographic names to foster communication and cooperation. So I think that fits really really nicely with the concepts of vocabaries and the work that we're trying to do. This group is probably only really a few years ago we started talking about linked open data. We very quickly acknowledged that it was sort of that next stage and a bit of a technical development for us and understanding that we needed to start to learn a little bit about what this meant. Most recently we hosted a webinar and we had some really interesting presenters from overseas. There's actually some from Norway, the Netherlands and Germany and actually demonstrating some really good working implementations of linked open data that we're using using place name vocabaries. The link is actually on the bottom of the screen though. So if you want to take a look at that please do. I think the preference though for our group is not to upskill ourselves into a whole data science space. We really want to encourage collaborations between national naming authorities and sort of linked data practitioners at that national and regional level. So thank you Steve and all the other organisers today for helping me in that little part. At an Australian level we don't really have a genuine, a single national names authority. We've actually got a couple that work together to do that job. So some of you might be familiar with the Australian New Zealand working group on place names. That's where the composite Gazetteer Australia comes from or the Geoscience Australia Gazetteer if any of you have used that and are familiar with that. It's also really, it's effectively a collection of all the state and territory and Commonwealth naming authority. So the government naming authorities across Australia and New Zealand and they come together to make decisions around what place names are going to be. So those official place names and they also make the most jurisdictional data sets. So it's a really a bit of a focus on that, that sort of decision making that authoritative kind of section of things. If we consider that working group on place names to be governmental or intergovernmental, the Australian National Place Names Survey which is the ANPS one is probably by comparison sits more in the academic space and the research space. It's a long running project to study and record the origin, the meaning and the motivation behind Australia's place names past and present. It was actually initiated by Australian Academy of Humanities quite a number of years ago but these days it actually sits as a not-for-profit. I think like a lot of things struggled a little bit with funding. So it's actually how now there's that place names Australia which probably no one has heard of even though it's got a really handy name for us is actually a not-for-profit. So that project and all that research in that space actually continues today with the support of a really small group of volunteer researchers, mostly retired academics which we're very grateful for. If anyone's used the TLC maps, the time laid culture map of Australia, the gazetteer that's behind that that's the data that comes from this group here. And finally down the bottom and as Megan mentioned my day job when I'm not down here having a chat to you guys or we're doing any of this other stuff is with the Queensland Government. So we're one of those principle government naming authorities that come together under that working group of place names and Queensland and the other states and territories as well work in various capacities with that Australian National Place Name Survey to share research across the sort of cultural aspects of names. So I think the purpose of trying to explain a little bit about I think is that it's to show that it's really quite an interconnected if not very well-known network of naming in Australia and that itself can make data of vocabulary creation and you reuse quite challenging. We've heard similar stories from a lot of people today. I think what's a little bit different about naming is that we're actually a very small community. So we may actually have a little bit of hope perhaps in getting on top of some of these things or they get too far away from us. Oh yeah. Yeah, I'll just give you that. Yeah. And we'll one second, we'll just switch. Okay. Is that okay about here coming through? Okay. Okay. Well, I'm sure someone will put something in the chat if it's not working for us. Yeah. Thanks very much. All right. So I'll jump now into some of the examples that I wanted to share and there are probably a lot that I could have talked about in this space, but I've chosen just three for you today. The first one I want to talk about place types and we've had a bit of a conversation briefly, I think in an earlier one about beacher types. So there's a little bit of an overlap there and a little bit of a space in there. When we talk about place names, it's a very, very broad term. We very rarely talk about it quite so broadly. We talk about things like hills and beaches and roads and towns and that type of thing. So our place names data is similar categorized based on the types of places and I should talk, I'm talking now really around that context of what the states and territories in Australia are doing and the data that then comes together to actually create a national gazetteer because you can see by the map there, we've got a whole bunch of states and territories. We've got some wet area. We've got external territories and we've also got Antarctic territory down there as well. Each one of them has a different data set. So it's a little bit of a challenge to try and create a national product not only for our own purposes but for anyone else to use for anything at all. The problem, we had a bit of a problem with this in that actually mashing all of that data together was a manual process. So there was a person sitting in Geoscience Australia, if you're lucky then, who had to source all of that data from the various jurisdictional databases, to do some kind of interpretation and join it all together somehow to create a product. And it was taking about two years to create a national data set and that was sort of cycling through all the time. So we had 10 data sets, of course it means we had 10 different sets of data types, sorry, place types. They weren't all completely wrong. We had had a national model for some time, a national vocabulary. It's just over the years that it sort of diverged a little bit. It was pretty similar, but it just not identical. That evolved a little bit. And of course there was a whole lot of variability interpretation because you had humans on the data creation end, sometimes putting place names into certain categories and then you had this person sitting in Geoscience Australia trying to mash them all together and sometimes putting them into different ones again. So it's a little bit challenging. This all came to a head. We knew it wasn't great, but it kind of kept going as is usually the case, the people who were creating the data and creating the problem weren't the ones trying to join it. So things kind of kept going along. And really what happened one day is the person who was sitting in Geoscience Australia doing all this work retired and his team, when they went to sort of distribute all that work, went, no, we're not doing this anymore. And basically, so to the states and territories, you guys got to put your heads together and come up with a vocabulary and some similar types so that we can just automatically do this. So that was our solution. We basically decided we're going to agree to a national vocabulary. Everyone was going to supply data using that. And we had this bit of an idea that there was probably a fair bit of overlap with topographic data and existing feature types. There was a whole range of different, different vocabs within that discipline and within government already. And we kind of mashed all those together to try and create a more modern one that worked for us without taking a huge deviation from our existing previous national model. So off we set. Very, very low tech approach here, very different to probably some of the things that you used to. We literally did some whiteboarding. We had a bit of some good ideas and then we sat around a laptop and created a bit of an Excel spreadsheet. So this is as simple as it needs to be and from a subject matter expertise within the discipline, we don't have any data, you know, the big data skills that many of you guys do. So this is where we got to. And it was actually quite reasonably effective. What we ended up with in that green column there is what ultimately became the place types and you can see they each had definitions. We didn't know anything about linked data. We didn't know anything about vocabularies. By pure chance though, we did a few helpful things through this journey. The green column there used to be two to four letter codes in the past. We had hundreds and hundreds of things. And so we decided this time around, we're going to do away with codes. There wasn't limitations in data storage like there was in the past. And it was hard for us to remember and look up extra code lists and things like that. I think you talked about some of that stuff before. So we got rid of those and decided to use whole words that a regular human probably had a pretty good, you know, handle on what they might actually be. We also created about, we created three different tiers. So you can see we've got on the file left, we've got different groups and categories and then all the features or the place types belong to those. Our thinking there was that, that was a little bit of just human logic. And we're also looking at ways to be able to provide different types of grouped products for people. So if they wanted to look at, you know, all the undersea features, which a particular organization might want to do, it was all in there together. And we also actually reuse some existing vocabularies. So for example, I mentioned the undersea features there. We didn't go down the path of trying to define all sorts of different place types for undersea. We just have fully adopted, there's an international standing committee on undersea feature names. So we just fully adopted their vocabulary and slotted it into our spreadsheet there. So I've learnt since then that some of those things are probably, were probably pretty helpful. It worked for our purposes. So the spreadsheet was agreed to, we came up with all of that. It was sent off the data supply, you know, states and territories went off and sorted out their data. Geoscience Australia built some new infrastructure. They put a cloud-based repository with a GUI front-end so that the state governments could load their own data. They built some scripts in FME behind the scenes that joined all those data sets together based on the feature types or the place types that were created. And then they built this new portal which is what you can see on the screen there. And you can see our place types and that sort of three-tier structure there is working through in the filter as well. So we were happy with that. That got our national data back up again and people could use that again. We didn't really know much about this, as I said already. But meanwhile, there was a lot happening around loci and Australian Government Link Data Working Group and of course Geoscience Australia people and I think there's a bunch of them online at the moment. We're working on this as well and they went, you beauty. We've got a national data set and a national vocabulary for place names which is quite fundamental to a lot of the other stuff that we need to do. So they went off and did their thing and we've now got our humble little spreadsheet that we've all agreed to at state and territory level sitting up there in the research vocabs Australia. So very low tech but I would say reasonably successful sort of situation that we ended up with there. So that's great. We've now got this data there. The second example I want to go into is well, what's next? So we've now got this, what is really still a very simplistic data set we don't have any status types. So these are things like is the name official or unofficial, is it local, is it indigenous? Is it a dual name? So all these terms that you're probably familiar with none of those were able to be sort of be able to classify at this point. We also obviously only had current data in there. So there's a lot of people wanted to wanted to know what the historical data was. Then we start thinking about what about proposed names and all sorts of things. All this information sits in the state and territory data sets. So it's there, but it was just a case that it wasn't able to be able to easily get up into this into this sort of space. So we had a similar problem starts to look very, very familiar. So we had our 10 data sets again. We've got all these 10 different types and things like that. So very similar situation to what we had before. And what was different here was a little bit interesting and this is generally actually no currency type. So there was this sort of blended bit of status sort of stuff. And I've given in that sort of mess on the screen there. These are just screenshots from four different examples of the status types from some of the state and territory data sets. So you can see there's a whole mismatch of approaches. You have to have a fairly good handle on how those jurisdictions work, what their legislation is. How their data has been managed and all of that sort of stuff to really start to make a little bit of sense of that. We haven't broached this yet. It is the next step in actually being able to create a useful national product for Australia in naming. But this is just something that I've sort of started to have a bit of a play around with. So the solution for this one. Is it the same as what we did in place types? Do we just get together again and go down that same path? Look, it could be a little bit simpler, simple, similar, but I hope not. I think a few years down the track, we know a lot more now about fair vocabularies. We've got a whole room and a whole cyberspace full of experts that can probably help us approach this in a, perhaps a smarter way than what we have done before and actually combine the subject matter knowledge from our group and our people around some of those real specifics and actually bring that forward a little bit more. I'm probably running a little bit short on time. So I'm just going to wrap up really finely. We're just one final example. And this one was important. I wanted to show you because it sort of flips things around and puts things on completely the other side of things. Because a lot of the problems we have at the moment is we've got multiple datasets with multiple different types of vocabularies. Here we've got something that we don't really have a lot of data for yet, but we arguably have a vocabulary. So what you're looking at there is basically a taxonomy that starts to classify the motivations of naming. Now, this is not really so much into the government space, but into that national place name survey space. And this is the core. This is bread and butter. Their work is really understanding the motivations behind different names. And if you've been paying attention to the world, this is a really interesting space for a lot of people at the moment. People are starting to ask, well, how many names are named after people? And of those, are they of good standing? So if you're sticking to things about things that are happening in like Black Lives Matters movements, gender equality, inclusion, diversity, all that type of thing, people are asking a lot of questions around this sort of stuff. Government authorities are starting to tag their data or put different categories in their data around how to record this. So they've had to put tick boxes if a place has been named after a woman. So they can start to do gender equality statistics and things like that. So we are at the beginning of that proliferation, yet we have researchers from this group here that have actually started already to do some of this work. So this is just an example I wanted to show. And I guess these are the questions that I put to groups like this. Are there vocabularies like this that are already out there that we can start to actually publish in a fair way which might help prevent proliferation of this sort of these kind of things and different vocabularies in the future? So just a whole bunch of food for thought. I won't go into that because I think I'm probably a little bit short on time and I want to let Megan sort of move forward. But I could probably fill a whole bunch of this. I think the thing that keeps coming back to me time and time again is the importance of connecting people and finding the right people in the organisations to connect together. This is not a data issue. It's not purely a technical issue. It's really a lot about people in that space. A lot of the work we've been doing has been about building vocabs and nontologies or models around data sets. And those data sets are often really simplistic. They don't really represent the real world. So I think there's scope for us to start to shift some of this thinking into actually modelling real world concepts or things like that and actually start to shift away from that and not actually have people continue to reuse the same old data that actually doesn't quite do the job. From my perspective, I think there's a lot of non-data benefits for subject matter experts. Some of the stuff I'm talking about here and some of these terms that need to be defined for data also exist in legislation. They define who has the power to make certain decisions and things like that. So there's a lot of value, I think, for us to start talking about outside of that space as well. There's obviously all the maintenance issues. I won't go into that. I think this group is probably well aware of all the challenges around that. But I guess my final question coming from the perspective that I do is how can builders of fair vocabs, which is a lot of you here, help guide how often some of these unfair vocabs or people like myself and the organisations that I represent do our work in a way that doesn't proliferate the same issues all the time at the end of the journey in that sense. So I'm just connecting that. So I might leave it there, Megan. And we'll go from there. Thanks.