 First of all, thanks to the chairs for the talk today. I apologize if there's a bit of background noise. There's a lot of workshops going on in the office today. And if you can only see me, I've turned my video off, so you can't see the positions that I'm standing in in order to be able to talk so that you can hear me. Maybe I'll turn it on for giggles later. So just a quick thank you to the chairs for inviting me to talk today. My name is Joe Aberatner. I'm the chief technology officer of a company called PSMA Australia. Just to give you a bit of background, PSMA Australia is an unlisted public company which is owned by the governments of Australia. So our goal is to facilitate broad and sustainable access to high quality location data. And so today's presentation is about code list government for one of those data products, which is the Geocoded National Address File. So I'll give you a quick intro as to what GNAP is. That's a Geocoded National Address File. To discuss some of the standards that are used in it, to describe the process for maintaining code list that currently goes on today. It's not much of an involved process. And that's why we're looking at describing a process or a proposed approaches to code list governance in the future, why the change will become evident as I go through the presentation. So simply put, GNAP is the trusted index of Australia's street addresses and their associated locations. I'll put a snippet in from our data product description which you can find online. I could probably put some links into the notes and I'll flip this slide deck through with some annotations later. So basically to go through where the data comes from, PSMA receives addresses from a number of trusted government sources. It runs a vast number of integration rules which is built up over time that essentially deduplicate the addresses. It identifies aliases or addresses known by other names and then checks them against reference data sets which include things like the roads, data set and the localities data set. It then links the addresses to other relevant data sets which include the ABS mesh blocks for linkages to the statistical data and Australia's land parcels in a product called Cadlight. Why is it important? I guess I'll take a stab at this in case people don't know why addresses are important. They're extensively used in navigation so most of you have probably used it to get probably not to your offices but if you're trying to get some petrol on the way home you may. It's commonly used in infrastructure planning. It's commonly used in government service delivery including emergency services although GNF's not used in the city of New York so the New York PD police picture is probably a little bit misplaced. As well as just the simple things like the post you're getting a mail to rural areas or a pizza delivery person getting delivery to your door or even just the estimates of how long it's gonna take before you get your pizza. It's so important that the government funded its releases open data in 2016 along with our administrative boundary status set. So that's available now for download from data.gov.au. As part of its commitment to improvement the Department of Prime Minister and Cabinet ran a survey to seek input for improvements from users of GNF and they received a number of requests for a standards compliant web accessible and queryable endpoint. So Pearson may have thought it could accomplish this through publication of the data set as linked data and went about publishing GNF and some administrative boundaries and connecting them to a similar publication task of the ABS we're considering. And so what we've looked to doing is publishing GNF publishing localities and local government areas from our administrative boundaries data set and then assisting in the publication of the Australian statistical geographic standard main structure and then linking the data sets all together. We started this activity with a whole bunch of other people, I wanted to call those out now. So there was a number of people who met in a workshop that we ran in Canberra with Geoscience Australia at the time and basically we had the data providers Australian Bureau of Stats and us, we had hosting service providers there which includes Syro and Geoscience at the time although things have changed a little bit. There were some people who would be building demonstrators as well as a whole bunch of different experts which again included Syro, members of the Australian Government Link Data Working Group, Rob Atkinson, the Centre for Spatial Data Infrastructures and Land Administration in Victoria and again, Ann's. And so part of that discussion was about getting some guidance on how we might go about this task. And we took some outcomes of that and then of course we've been working with Syro, Nick Carr in particular to get that data published. A big thanks to you, Nick and everyone that's had an involvement in that. So talking about standards and linking it to code lists comes in this slide where we sort of have a set of standards that we comply to. One of these is a standard called AS 4590. So that relates to the exchange, interchange of client information. Quite a lot of you are probably aware of that standard. The rest is just one of the components that are covered in that. The standard defines the set of components of an address which includes things like flat type and number, level type and number, road number, the things that you see on the screen there. It also includes Geo code and its types which include things like front centre setbacks and parcel and property centroids. Unsurprisingly, a few of these are listed as code lists and the standard goes on to in appendix C describe a code list for flat types and level types in what it calls addressing abbreviations and D appendix D covers off road types and road suffix abbreviations. And so it provides the standard does show code lists for all those types. So the governance of those code lists is quite limited. AS 4590 is regularly reviewed or hasn't been reviewed in a while. I think it was actually reviewed but I think something's been released yet unless others in the room know. PSMA is often asked to provide its codes into that because the standard should be covering legacy in use codes and PSMA because of its role in the addressing supply chain is seen as a source of the in use address abbreviations for things like the street types, road types, sorry street types, street suffixes, et cetera. They're largely accepted when we hand them through but the problem is it's a long time between drinks I guess on the review side. And so we have to still keep these things are in use, people still want to have access to the codes for things like input validation and so it causes some problems when there's a long time between drinks. So looking at a better way of doing this because we're trying to publish these things, this link data, there's going to be a whole bunch of codes that are packaged up with the data and we want better ways of managing those codes. So that's why we're looking at this now. This is a slide from our colleagues at Syro that they've put into a recommendation for a farmer's data marketplace. And essentially they've been looking at the use of social architecture as a means of deliberately designing the institutional arrangements that underpin information infrastructures for quite some time now. So I borrowed this image from this document that I've referenced. And I want to zero in and bring your attention to the area which is highlighted which is around, it's a section called rules of the game. So borrowing further from their insights, we've proposed to manage the code lists as registers and the relevant standard for that is ISO 19135 which is the procedures for item registration. We're using that as a guide to identify things like a list of roles for people who would be involved inside the governance tasks that we've got ahead of us. So the main roles that are in there are register owner, control body, submitting organisations, register manager and registry manager. But the question is who would play those roles and that would be a common question for a lot of these things which are sort of community managed. Here's an example of what those roles mean. I borrowed this from the Indigenous Locations working group which was set up quite some years ago. I'm not sure where this is up to nowadays. I might throw that to the group. Is anyone aware of what's happened to this? Because I was very... But I haven't heard anything in a while. Joe, do you mean specifically what's happened if anything to the Indigenous Locations working group? Yeah. I can't say I know. Let's wait and see if anyone puts anyone in the question. Sure. So ultimately, we've got... I've spoken about a whole bunch of different roles. If you see at the top, you've got a register owner who then appoints a control body. They delegate the... So they effectively have the decision authority. You then have someone who's managing the register and they request decisions from the control body in order to make sure that the content's kept up to date. They then make the edits to the register. They may choose a separate registry manager to effectively host or store that register, make it accessible. And then that's used by the users who I haven't covered in that previous slide. Has anyone got any questions there? Has everyone sort of seen this standard before? Joe, on... I'm seeing what I've got, I think some of us definitely have, but yes, I mean, it was intention to get you to talk about, you know, real local deployment of this, and yes, your question is, has anyone else used it? So I thought to say this before, if people are interested in either commenting on what Joe said or asking questions, there's a chat box here. Just put your questions down in that chat now, and then at the end of Joe's talk, we can visit those and, of course, for Joe, and for Bridget's talk and see what's written there. OK, I'll just push that up so I can make an answer if they're coming in. And there's probably others who could too. So I'll just move forward. So it turns out, in the world of addressing, these are actually fairly clearly defined. So in terms of the register owner, we've got the Intergovernmental Committee on Surveying and Mapping, and they're the people who have had the most input into these codelists when AS 4590 comes up for review. So despite the fact that they're not the main people who are reviewing the standard, I think that's because it's such a broad standard. But when it comes to what are the codelists that are being used? They've historically used previous standards created or managed by the Intergovernmental Committee on Surveying and Mapping, or ICSM. But some of those standards, the codelists have been reduced because the main standard that they use is one about new addresses. And of course, that doesn't consider some of the legacy ones. They want to try and reduce the number of ones that are used for new addresses. So that list has come down to 15, I think it's 15, or a very small number of codes in the codelist, but that's not the whole in-use thing. The in-use thing now is sort of being sought from PSMA because we have the largest considered most complete data set of those. So the control body was there's a permanent committee on addressing which sits inside the ICSM structure and they're a logical place to delegate the decisions with respect to addressing. Submitting of organizations essentially, for us, if GNAF becomes the register, then the people who already supply us make sense to be the first set of submitting organizations. If that list then needs to be increased, then we can go about doing that. Then PSMA is the register manager. Once we've got a better handle on how we want to go about this publication task, then the idea of who would be the register manager will become more clear. At the moment, Syro, again, another thank you to Syro who are currently hosting the link data product that we're working on, and which currently contains those codelists. During the course of the link data demonstrator, we identified how a business process for maintaining the codelist might work. We've proposed the business process to the nominated registry owner, which is ICSM, who have accepted their role and delegated to the nominated, sorry, and the delegation to the nominated control body, which is a permanent committee on addressing. We're currently waiting for the decision to be documented in the permanent committee in addressing minutes. And in fact, I think they need to accept that responsibility as part of that process as well. I don't believe they've met since we've asked them to accept their responsibility in this governance process. The GNF contributors will become the first submitting organisations. The business process allows for the resource to be live and still have solid governance. This is done by effectively having an experimental data resource that we manage statuses according to ISO 19135's statuses. I've added the proposed governance process to the Australian Government Link Data Working Group Github repository. So I'll include the link to that below. So if anyone wants to have a look at that, you can just click on that link and get in there and have a look. And I guess what I'm keen to hear are people's thoughts, comments and feedback on that. We're looking to hold a retrospective, which what I'd be doing now is inviting those people who participate, all those organisations who participated in the first workshop to get back together to present to them what we've delivered in terms of the link data resources, what we haven't, why, some of the impediments and discuss some of those things that get in the way of this task so that we can try and learn from the experience. And then when we start to plan next steps and we still have a little bit of time left to accomplish the entire lot of things, we can learn from what we've done so far and the collective knowledge at the table. Happy to keep you guys informed of that. So if you put any feedback or comments into that retrospective process, then I'll bring it to the table when that, and there's a couple of other people in the room. I noticed that it also liked to have who were also there at that first workshop. So any one of us can bring your comments into that room and then keep you informed of the status of those recommendations in the actions that are taken after that. So yeah, so that's it for me. Get involved. We're doing a lot of our knowledge sharing information updates to the Australian government data weaponry. There's a link to the in-progress data set. That's sorry, let me just turn the volume down a little bit. There's a link to the in-progress publication data set. Sorry, yeah, data set that we're publishing over there. Also, if you're going to make an email or anything like that, then I'll put it in my email. Contact details. I'll annotate these presentations too.