Professor Marcia Langton, Dr James Rose and Dr Len Smith discuss Digital Curation in the Indigenous Data Network.
Based at the University of Melbourne, the Indigenous Data Network was created in 2017. Its dual purpose is to strengthen local Indigenous communities’ agency in managing their data and to strengthen policy and decision making.
The Network aims to archive orphan data sources and increase visibility of existing data; work with government agencies and non-government organisations to ensure Indigenous communities’ access and ownership of data; ensure that official data collection reflect Indigenous priorities; and coordinate educational programs to ensure Aboriginal and Torres Strait Islander people have expertise in data science.
Joint event with Digital Preservation and AIATSIS.
Digital Curation AIATSIS Event
*Speakers: Marie-Louise Ayres (M), Craig Ritchie (C), Marcia Langton (L), James Rose (J), Len Smith (S)
*Location: National Library of Australia
*Date: 26 June 2019
M: Good morning, everybody and thank you for being patient. We’re starting a little late this morning. Now I don’t know why it is this morning because many of you will have enjoyed a beautiful rainbow over Canberra but lots and lots of things have gone wrong this morning in terms of who was going to be here at certain points so there’s just been a little bit of regrouping to make sure that we can hold this wonderful event today.
So firstly I wanted to welcome you to the National Library of Australia. I’m Marie-Louise Ayres, I’m the Director-General of the Library and I really wanted to begin our day by saying that you are meeting here today on a building that stands on the lands of the Ngunnawal and Ngambri peoples but also I think an organisation and a culture that is increasingly conscious that it is standing on the culture, the custodianship, the stewardship of knowledge, country, culture that has gone on in this place for a very, very long time. So I honour the elders of the Ngunnawal and Ngambri peoples past, present and emerging and particularly thank them for the generosity that they have shown and are showing us as an institution and sister institutions in sharing culture, knowledge and understanding, especially in this year, the International Year of Indigenous Languages.
I also really wanted to pay my respects and deeply thank my indigenous colleagues here at the Library – many of you are here in the audience today – whose work, words, thoughts and deeds are profoundly changing the ways in which we think about our collections, about data and about our mission to serve. I just really wanted to say that I’ve worked in collections my entire career and I have long known that our collections here at the Library contained deep knowledge that could be much more useful to indigenous communities around Australia and we’re certainly working on that. But after a lifetime in working with collections over the last year I’ve felt these weird shifts going on in my brain where I realised that the things that I have thought of as archival collections that have their being because somebody kind of created them are actually just kind of a cypher, just a surface over a kind of deep culture and knowledge that is just trying to break through out of that paper.
So my understanding about what our collections in this space has has – it’s like taking – I don’t know, a complete shift in how you think about your collections and think more about the kind of mini voices that are underneath them just trying to get out. I think this is one of the things that we really need to think about in terms of digital curation and around indigenous data, is to think what are we talking about here? Whose data is it? Whose knowledge and culture? Whose people? Whose identities is it? In this data and how do we think about that in ways that are a little bit different from our traditional custodianship and stewardship roles.
So I say vive la difference and that we should continue to be willing to be challenged in terms of thinking what is the task that’s ahead of us as we try to curate data in this particular space – different things shifting. So I hope this morning is a day where you might also feel those profound shifts where all of a sudden you’re thrown off balance after 25 years in the business and think gosh, I know nothing. They’re great moments. So I hope you’re going to get that kind of discussion this morning and it’s going to be led by my colleague, Dr Craig – almost, not quite doctor. He’s working on the doctorate, I’ve seen the long hand. I think I’m so overwhelmed at the thought that my opposite number at AIATSIS, Craig Ritchie, is both running his organisation and just doing a PhD in his spare time. I really don’t get this. But Craig of course is going to basically draw out these conversations between our speakers today, some in person and one we hope virtually so Craig, over to you. Welcome again, remember whose land you are standing and sitting on today. Thank you.
C: [Speaks in language] 5:04. Well good morning, everybody. I’m Craig Ritchie, I’m a Djangadi man. That was me dazzling you with our Djangadi language. For anyone who speaks it, probably adulterating it. I’m a Djangadi man, our country is Kempsey on the mid north coast of New South Wales from the mountains from the New England Ranges into the sea. Can I join Marie-Louise in acknowledging that we’re on the lands of the Ngunnawal and Ngambri people? Pay my respects to their elders past and present. Can I extend that respect to Aboriginal and Torres Strait Islander people that are here with us today and all of you, thank you for coming along today.
As you heard I’m the CEO of the Australian Institute of Aboriginal and Torres Strait Islander Studies, my view one of the – but an institution that has had and is positioned I think to have profound impact on the way that our country grows and develops.
Today’s event of course is a collaboration between AIATSIS staff, National Library of Australia staff and Aus Preserves, a collaboration for which I’m deeply appreciative and want to thank all of the staff that are involved.
As you heard we have three speakers, two of whom you can see, one of whom we hope you will be able to see online. We also have various ways to enable people to participate in today’s event. Obviously if you are here in the theatre as we get to the sort of panel discussions, where I will morph from CEO into either some combination of David Frost, Michael Parkinson or Graeme Norton and I’ll leave you to determine which of those three, you’ll be able to participate obviously in person. If you’re here but you feel uncomfortable about standing and asking a question there's online options. Now the fact that I’ve stood up with a pen and paper will give you some clue that I have had some assistance in navigating the online options and we have an ability to field questions online. My colleague, Anthony, over here is going to assist me with that. But we want to make sure that you’re able to get a hearing and that you’re able to ask the kind of questions that you need to ask.
This forum is a vital one, it’s a vital interest to the Institute for a couple of reasons. Firstly we have a particular responsibility at AIATSIS, we feel a particular burden to make sure that the management of our collection in particular pays due regard to the concerns and aspirations of Australia’s first people. It is our cultures and it is our identities that are intimately concerned here. So the indigenous data network’s interrogation of the processes around data collection, its management, its discovery, its access and its use bears directly on questions that are fundamental to how we at AIATSIS go about our business.
Our vision at the Institute is a global one. The first few words of our vision statement are we see a world and it’s a particular world, it’s a world in which Aboriginal and Torres Strait Islanders’ cultures, knowledges are recognised, respected, celebrated and valued. It’s of prime importance to the Institute, to the Council and indeed to all of our member and stakeholders because this is critical to the cultural resurgence of Aboriginal and Torres Strait Islander people and all of our staff wholeheartedly supporting that vision by making sure that we care for the collection that we have, we make it accessible and in ways that respect and honour the people who own that knowledge and that – and especially for Aboriginal and Torres Strait Islander researchers.
Although we’re in a bit of a peculiar situation. We’re 55 years old this year, the first Act that established what’s now AIATSIS passed the Parliament in 1964 and in many ways as a statutory body we are a direct product of both colonisation and dispossession. Fifty-five white academics, largely male felt an impetus to do something to document and preserve Aboriginal cultures as they understood them in 1964 before as they expected we disappeared from the pages of history which manifestly we have failed to do and we aim to continue that failure in particular.
But if they hadn’t felt that burden, if they hadn’t felt that drive and if that academic drive hadn’t been matched by the energy of what some might regard as one of the oldest members of Parliament to ever sit in the House of Representatives, Billy Wentworth, the Institute that I lead and that many of us in this room work in wouldn’t exist.
Now since 1964 the Institute has evolved to become not a backward-looking institution trying to document a culture that was disappearing but as a forward looking, proactive institution that seeks to tell our stories and give people the opportunity to engage with those stories, to support the cultural resurgence of Aboriginal and Torres Strait Islander people and to influence the kind of story that this country tells itself about itself and that it tells the world about itself. We have – one of our core commitments and our organisational values that we are striving to unpack and think about and operationalise is how in everything we do we have an Aboriginal and Torres Strait Islander voice or voices being heard.
So we have to keep working at best practice and working and learning from our peers and so collaborations and events like this are critical for us because we want to hear from you, we want to hear from experts in the particular field under consideration today so that we can learn together. Now as I said we have three speakers. Our first speaker probably needs no introduction but since she’s not here it’s probably necessary even more to speak a little bit about her.
She of course has a distinguished academic record and a proven – and has a proven record as a fearless – I’m not sure that adjective does justice – fearless and articulate campaigner and commentator. Many of us at the Institute feel that the 2018 NAIDOC slogan, Because of Her We Can, applies particularly to Professor Marcia Langton who was the first woman to chair our council until this year when - before the selection the Minister appointed Jodie Sizer as Chair of the AITSIS Council.
Professor Langton was described along with Pat O’Shane as a fierce, passionate warrior who has shown that we can go places where other Aboriginal people may not have yet ventured and indeed where many non-Aboriginal people thought we couldn’t. Rather than reciting her bio which you all have I will simply introduce Professor Marcia Langton. She’s the Founder of the Indigenous Data Network and today she will speak about its inception at the University of Melbourne in the context of the global indigenous data sovereignty movement and indigenous science, technology, engineering and mathematics. Would you please make her welcome through the ether?
C: Hi Marcia, over to you. Make her welcome again.
L: Thank you. Hello. Thank you. Shall I acknowledge the traditional owners? The Ngambri people where your conference is being held and I acknowledge the traditional owners here, the Wiradjuri people of the [Coolum] 15:58 nations.
In 2017 the Indigenous Studies Unit where I’m located at the University of Melbourne in the School of Population and Global Health organised the Indigenous Data Sovereignty Symposium with the specific aim of developing a nationwide network to empower indigenous [unclear] 16:24 developments in data science and maximise the use of their data resources for community benefit. We also aim to increase awareness of the importance of indigenous data sovereignty for local indigenous communities, researchers, government and other related stakeholders and provide information on custodianship, management of and reporting of data, presentation of data including models of monitoring and evaluation.
Following that symposium we established the network because of the overwhelming agreement that there was a need for a network that brings together experts from universities, institutions, governments and communities. Our partner at that symposium was the Australian Institute of Aboriginal and Torres Strait Islander Studies, also the [Lowecher] 17:34 Institute and we received a great deal of support from the University of Melbourne.
Since then we’ve established the network and you’ll hear from James Rowe soon about the importance – the ideas that we are proposing to our network. So one of the things that we found particularly important during the symposium and since is the need to partner with communities who are attempting to establish their own data units. Our first partnership is with the [Algabunya] 18:22 Data Unit in Shepparton and with [Algabunya] and the assistance of the Australian Institute of Health and Welfare and other agencies such as AIATSIS we are identifying best practice in community data collection, applying best practice in data management. We’ve already formed a technical subcommittee of our steering group. We have designed a survey for identifying those indigenous datasets that exist. We have for instance already identified four orphaned datasets with a view to integrating and archiving indigenous datasets and preventing the orphaning of important datasets which would be detrimental to communities.
We are working with the Indigenous Research Exchange to develop guidelines and best practice case studies for research and data analysis and we are developing educational programs. So the steering group consists of the Chair, Professor Sandra Eades in the School of Population and Global Health at the University of Melbourne, Professor Shaun Ewen, Director of the Poche Centre for Indigenous Health, Associate Professor James Ward from the South Australian Health and Medical Research Institute, Craig [Kalinga-Griffiths] 20:21 - Dr [Kalinga-Griffiths] at the University of New South Wales, Professor Janet McCalman and Dr Lyndon Ormond-Parker and Dr Christian Smith from the University of Melbourne, and Len Smith who’s there with you today.
That’s enough from me for the moment as an introduction to the Indigenous Data Network and our rapidly developing program of works.
C: Thank you, Marcia, for that introduction. Our next speaker is Dr James Rose. He’s a senior research fellow with the Melbourne University School of Population and Global Health in the Indigenous Studies Unit where he is the Co-ordinator of the Indigenous Data Network. Prior to his appointment he had long experience in indigenous communities and organisations developing information systems that supported their programs and objectives. Dr Rose will explain the technical basis on which the network operates and including key terms, definition, strategy. Thank you. Would you give him a welcome?
J: Thanks, everyone. Before I start let me reiterate our thanks to the Ngunawal people. In my previous incarnation as a forensic social anthropologist working for a native title representative body in Redfern I helped conduct quite a bit of research in this region for a prospective native title claim. The last claim that I worked on was just to the east of here on the south coast. It’s been documented in the media so if you’re interested in the kind of work that generates data then that would be a good place to start.
The - you’ve heard the introduction from Professor Langton. The way we’re going to roll from here is that I'm going to go through the technical terms and definitions that underpin the work we’re doing at the IDN and the standardised language that we want to build nationally to begin creating a consistent discussion around policy in this area and then I’ll be followed by Dr Len Smith who’s going to talk a bit more about the specific development of the IDN.
If you have a Twitter account please follow us on data indigenous. Okay, there we are. Now am I controlling the slides or is – okay. Great. The first term and definition that we have to talk about and – which is very hard to find literature on, especially in this particular space is that of data. What do we mean by data? Particularly what do we mean by indigenous data in this context when we’re talking about curation?
I’m going to rewind a little bit into my anthropological and linguistic training and talk about where data sits in a development chain, okay, a semantic development chain that all people participate in, all cultures have this kind of production process which starts with human experience of the world at a physiological and neurological level.
What happens when each of us moves through space and time and interact with another is that we detect what are effectively changes, physical changes in our environment, changes in the electromagnetic environment which are converted by our sensory systems into – get to data, eventually. What this means is that each of us has a universal set of equipment for detecting sound as changes in air pressure, changes in electromagnetism which we detect as sense and changes in the wavelength of light which we call sight. What happens when those changes detected by our sensory equipment is that it’s processed in our brain at a subconscious level and cognised so that we can react in different subconscious ways according to the immediacy of those detection.
What makes human beings so special evolutionarily is that we have developed a very sophisticated capacity for symbolic construal or realisation of that experience in the form of language. So other species have limited forms of language but homosapiens and possibly other [unclear] 25:07 for language and what language allows us to do is realise meaning, we generate meaning from our experience. So on the one hand we’re moving through space and time, we’re interacting with each other, our brains are processing changes in our physical, chemical and electrical environment and then we’re cognising that just like all the other animals that we share the planet with but what we do with language is generate a very elaborate layered system for turning that experience into meaning.
Without going into too much detail what we get out of that is information. Information is a universal product of human experience using language to create meaning. All cultures generate information. Information is the specific domains of experience that have been transformed using specialised domains of language, specialised types of language. So in the Academy we talk about faculties and departments and schools and so forth which reflect increasingly specific branches of language for realising meaning which allow us to do increasingly effective things to the world around us and to each other, be it in medicine or engineering, computing and biology or whatever. Literature, art, music, it’s the same.
[What’s] 26:35 specialised those areas of language sufficiently is that we are able to quantify elements of those languages into characteristic units, characteristic traits which we can assemble into collections. It is those characteristic traits that we refer to as data. Data are the characteristic attributes of already specialised domains of language which we call information that we can count. That’s what data is. It’s the same in every single field so when you’re creating a spreadsheet and you have columns what you’re doing is distinguishing characteristic attributes from the field of language that is already specialised that you’re using to create meaning about a specific domain of experience.
Okay, we all do that. Even outside of the Academy we do that. We do it when we learn [unclear] 27:30 laces, for example. We learn to go to the toilet as small children. I can say having had small kids that’s quite a – elaborate process. At the end of it if we turn that into a specialised field that can be reproduced in a controlled environment like the Academy then we get data that we can count.
So what’s very important to this conversation about indigenous data is the concept of property. The World Economic Forum in 2011 released a paper in response to burgeoning pressure from Web 2.0 participants. When I talk about Web 2.0 I’m talking about the dominant media platforms of the internet with which we engage actively. Social media, for example, Amazon, eBay. These are not static articles on pages that we used to read back in the ‘90s, that was Web 1. Web 2.0 is a participatory environment and people are contributing incredible volumes of information to the platforms from which they need to receive a service.
The World Economic Forum in 2011 said that this economic activity, which is what it is, does in fact constitute a form of intellectual work on the part of people who are participating on the internet as clients rather than service providers and as such it generates an asset, it generates actual value. So they describe this as an emergent asset class, a new asset class. Data, personal data specifically which is comparable to tangible assets such as real estate and infrastructure.
Then around about the same time the International Accounting Standards Board and the IFRS, the International Financial Reporting Standards Foundation consolidated their standards to reflect this refined definition of data as a form of property generated by work to fall into line.
So that now before we go any further last year the European Union I think related the GDPR, the General Data Protection Regulation which explicitly describes the data that is held by Web 2.0 platforms such as Facebook, Twitter and others as the property of the people who generate it and that each European member nation protection authority as a consequence to whom individuals can report breaches of that regulation and request that all their data be deleted. You might have heard of this in terms of the right to be forgotten which that bit of legislation now enshrines in the European Union.
We don’t have that in Australia, we have layers of privacy regulation but it’s not the same thing, especially when we’re talking about indigenous data. What is indigenous data? The concept of indigeneity is the product, the inherent product of a colonial exercise such as that that was undertaken in Australia otherwise there would just be Australians. What happens – what has happened in Australia can be characterised as an organised unilateral asset transfer from indigenous people in the first instance to Britain and after the devolution of governance by Britain to the Australian Commonwealth thereby to the Commonwealth which now resides primarily in this city.
The staggered nature of that asset transfer started off with land, the alienation of land in a wholesale act of appropriation of the continent of Australia. It was then followed quickly by the alienation of natural resources, alienation meaning of course that people who originally enjoyed access and use of those resources – alienation means to be locked out of. It was followed then - the alienation of natural resources was followed by the alienation of labour which we know as stolen wages or slavery in other parts of the world. Then as the colonial administration developed and became more sophisticated by the end of the 19th century in this country we begun to see the development of a specialised language.
So I spoke earlier about the development of specialised languages such as biology and chemistry and so forth. In the colonial administration one of the most important forms of specialised language is that which regulates indigenous people, indigenous labour and indigenous-owned assets and we know this today variously as ethnography or anthropology or any of its affiliate branches, specialisation. So these are specialised languages that have specialised terms and definitions and a specialised way of indigenous labour, indigenous land, region, ceremonial practices, economies and so forth which seeks to regulate it. So it turns it into a domain of knowledge for nonindigenous people that can be controlled. Out of this highly refined and distilled formal language of course we get data and this is indigenous data, this is how we arrive at the point we’re at today.
To contextualise a little bit, everyone’s very familiar with the concept of land rights and that was an attempt to restore or reverse some of the alienation of the land that was initially taken by Britain and we’ve had an attempt to reclaim some of the stolen wages so we have begun to address as a nation some of the issues around stolen labour. Now we’re beginning a slow process of beginning to try and restore the property right around the appropriate indigenous data. Sitting in between labour and data we have information - this is another area of law which is covered by IP trademarks, copyright and so forth. That’s quite well developed but law around the restoration of indigenous data is not well developed and that’s what we are aiming to address with the Indigenous Data Network.
Len is going to follow shortly and discuss the more specific applied features of that endeavour so without further ado I’ll hand over to him. Thank you very much for listening.
C: Well there will be a bit further ado before Len gets up and that is a brief introduction. Thank you, James, for that, lots to think about in that short presentation.
So Dr Len Smith is an Associate Professor in the ANU School of Demography, a visiting senior research fellow at the Indigenous Studies Unit, Melbourne School of Population and Global Health. He has conducted considerable amounts of research in indigenous epidemiology and demography and as you heard today he will – well according to my notes anyway – discuss the importance of linking orphan datasets to enhanced policy development in government service and probably anything else he wants to talk about. Len, thank you. Give him a welcome.
S: Thank you, Craig, and I join my colleagues in acknowledging the Ngunnawal people on whose land we meet. I’m hoping that the balance of slides will come.
Well as Marcia said she convened a National Indigenous Data Sovereignty symposium in 2017 and the background to that was not only the international movement in data sovereignty but also a series of Australian initiatives funded by the federal government to establish what were called empowered communities selected for their representativeness around the country and part of the development of the empowered communities was the development of data capacity within them. Some of those empowered communities have really forged ahead including the community at Shepparton that James mentioned. Others have not developed so quickly and so part of the idea of the network was to take the lessons from those that have succeeded and try and generalise those so that they could be used by other communities, not just the empowered communities but any other indigenous groups who are interested in documenting their own situation and using it – using that information for their own.
The second component of the network that was identified as a priority was allowing or facilitating the access to government data by indigenous communities and organisations. A vast amount of information that’s collected by government agencies about indigenous people and communities and often it’s very difficult for those individuals and communities to access the information about themselves. So there was – it was recognised that there was a need to free up the flow of information back to the people so that they could document their own situation in the terms that governments used to and use it for their own development.
Finally the third priority that was identified was the sort of rescuing of a vast array of indigenous datasets which we’ve – they were once well and established within a family network but through the passage of time were neglected and were in danger of being lost and there’s a very large number of those generated by government, by indigenous organisations themselves and by researchers.
So the idea was that we should get a picture of what’s going on in these three priority areas, get some resources and start to move on all fronts at once. An important component of that was the idea of capacity development, that this had to be an indigenous empowerment process as well as anything else so we want to see this network generating a [unclear] 38:20 of indigenous data scientists.
As Marcia said the steering committee was set up and we’ve now set up a technical advisory group to help us with the process. Speaking in very general terms we thought that the priorities can be defined as being about – on people, on country and on language and culture. Of course there’s a range of challenges. This institution that we’re meeting in is one of the components of the sort of network of organisations that need to be marshalled in order to address that and everything we conceive of is in terms of a partnership not only with NLA but especially with AIATSIS but a whole range of other government and non-government organisations as well.
What’s the aim? The aim is to preserve the data, to make it generally accessible and to allow it to be reused, to be shared and to be integrated.
Well over those priorities that I mentioned the – I think it’s important to stress that the empowered communities component is already a demonstrated success, especially in the area of restorative justice but also in school attendance and so on. Communities have demonstrated their capacity to use information in order to deal with priority issues that they themselves have identified.
Well to get down to real specifics what we want to do is create – establish and develop relationships with the key organisations in the government so that indigenous organisations and communities can access information from the census which is the main source of information about indigenous people but also all those other huge official data collections which contain information about indigenous people and communities like Centrelink, Medicare, hospitals, births, deaths and marriages, etc, etc, etc. Now of course there are issues of ownership, of confidentiality and so on, particularly when you address the issue of indigenous access and researcher access but that’s why we’ve got this network being set up, so that we can address those issues.
Without spending too much time on it my own sort of main concern is about orphan datasets and I know it’s Marcia’s. Since data started being collected by Europeans people have documented indigenous genealogies and family histories, some of those records are well developed, some of them are very poorly developed, many of them are in danger of being lost. They need to be rescued and put on a sustainable footing. Community surveys, same thing applies. Countless surveys have been undertaken of indigenous communities, where is all that data? Well we know where some of it is, it all needs to be documented and drawn together so that people have access to the information about the history of their own communities and organisations, health surveys.
A particular obsession I have is with the National Trachoma and Eye Health program which Fred Hollows ran. They have [unclear] 42:43 over 100,000 people. That information is sitting largely in paper in a basement somewhere and needs to be rescued and used as a baseline for contemporary research. Land, water, environment, crucial to the whole process of native title and land rights and so on. I could spend the whole day talking about all of the orphaned datasets that we need to identify, document, put on a sustainable footing and make available to communities and organisations.
Well the vision we have is one of full spectrum curation which means the – all the high-value datasets that we’re interested in are a mixture of pieces of paper and old media, CDs and data tapes and so on and very up-to-date databases as well. So it’s not just a data science question, it’s a – archiving question, it’s a curation question in the broadest sense of the word with the idea that digitisation of machine readability but you end up with something that’s sharable by the most modern means we have, usually via some sort of web interface.
I just want to say that we had high hopes that the federal government’s increase process, the collaborative research infrastructure process might provide the answer to all this because they were addressing this issue in the broad in terms of digital humanities and social sciences which they regarded as covering indigenous data. That process has stalled for a variety of reasons and it’s our belief that the Indigenous Data Network in a way can leapfrog that because the key to success in other areas has been not just the sort of partnerships I’ve been talking about so far, communities and so on but also a partnership between sort of people who are involved in the subject matter and the computer scientists.
The two successes that we’ve seen in this sort of partnership are firstly, very successful London group that has been involved in recreating the Anglo Saxon history of the UK essentially but also we look at our colleagues in genomics and the difference between them and the social scientists is this intimate partnership between the subject matter specialists, the archivist and the computer scientists and that’s what we’re trying to turn this network into. So that’s about it. Thanks, Craig.
C: Well thank you, Len. Thank you for our three speakers. So we now move to questions and we have roving mikes, we have a couple of mikes for our two folks down the front and can I just remind everybody that as this event is being – wait until you have a microphone before you speak so that you’ll be heard but also captured and – forever and forever. It would be great if you could identify yourself briefly when you ask the questions and as I said before there is an opportunity through the Facebook comments for you to be seen in your questions and they’ll be relayed to the tablet down here through they mysteries of the ether and we’ll be able to make sure you get those questions asked. I can see Marcia here, you can’t see Marcia but – can you hear us, Marcia? I can’t hear you. You might need to unmute your microphone. There you go. Great, alright. So let me just slip over here and do we have an opening bid for a question? Way up the back if you could just – I know you’re shy but if you could stand up and identify yourself and –
A: Hello, I’m Bronwen Cooper from AIATSIS, thank you all very much, you’ve answered a lot of my questions about exactly what indigenous data is and where organisations like ours have a part to play. You’ve talked a bit about learning from successes and I'm just wondering if you could help me to visualise that by unpacking a bit around the Shepparton example?
C: Is that question to anyone in particular?
A: Well I think a number of people talked about it, James, I think in particular, you mentioned it.
C: The bit about the Shepparton project, Marcia.
L: Yeah, thanks. Good question. So the Kaiela Institute in Shepparton is the lead organisation in the empowered communities network and the region around Shepparton is an empowered communities region. They’ve been working for some years with the other leaders from the other empowered communities regions to develop an approach that involves local community organisations and local Aboriginal and Torres Strait Islander populations in planning and prioritising policy – policies, funding and implementation of programs in their regions. In order to do so Shepparton – sorry, the Kaiela Institute in Shepparton has developed the [Algabunya] Data Unit and other empowered communities regions are doing likewise so there’s one in Kununurra also in – and the East Kimberly is also one of the empowered communities regions.
The point of having local data units is to – well first of all to collect their own data so for instance by doing household surveys, community surveys, having consultations with community to develop a view of local priorities. So in the Goulburn Valley region around Shepparton one of the local priorities is early childhood education and that is one of the 56 priorities that the community has developed and they’ve worked with an epidemiologist and others to combine their local data with existing data from the ABS and so on to get a picture of their own region.
So I think when I realised – when I realised the community report card, I thought well this looks to me like indigenous data sovereignty in practice, developed locally without any particular knowledge of indigenous data sovereignty but an organic movement to empower local people and to develop priorities and to collect, use and interpret data for local priorities. So that is an empowerment process and we’ve partnered with them in order to assist them with technical advice and to further develop their model and also to work with the other empowered communities regions to formalise and perhaps make more uniform a sort of national Indigenous Data Network of local community data units like the [Algabunya] Data Unit. Does that answer your question?
J: Marcia, I might follow that up with a supplementary. Do you think what we’re trying to achieve or – I think it’s interesting that this has sort of begun with empowered – A, of framing interaction and relationship with those communities. If there wasn’t an empowered communities sort of framework in existence and there wasn’t in the context of that some regional or local institutional capability would the sort of things we’re trying to achieve be possible?
L: Well it was – discovering the [Algabunya] Data Unit arose from the University of Melbourne partnership with the Goulburn Valley region and the University of Melbourne contribution to the development of the Kaiela Institute, the Academy of Sports and Health Education and the work that we’re doing to ensure that local Yorta Yorta people and other Aboriginal people in the area enrol in award courses at the University.
When I sent out the invitation to the data symposium the [Algabunya] Data Unit put their hand up and I realised that a paper from them would be an excellent demons – of the needs of communities but as it happened the symposium was held before the actual development of other data units in the country. But through our network we knew of other programs and they’re not involved in the empowered communities network at all so for instance my colleague, Dr Lyndon Ormond-Parker, has been working at [Wadaya] 53:56 for many years in the Northern Territory where he and his team have digitised 40 years of community audio, video, photographs much of which concerns their cultural life, their church life, their funerals, their sporting events, community events and he set up a television intranet which is broadcast so that the digitised community archive can be broadcast on television screens in the community.
So people from [Wadaya] came to present because that is another case of data sovereignty in practice. So they received all of their master tapes back from the Institute. The Institute – the Australian Institute of Aboriginal and Torres Strait Islander Studies archives the master tapes and that’s quite a different scenario from the Shepparton scenario. We also had the project from the Yawuru people in Broome so the Yawuru have a native title determination and agreements and a very sophisticated native title corporation. The Yawuru people wanted to do their own household survey and they partnered with Mandy here at the ANU and they presented on their household survey and why they were doing their household survey is so that - they want to have their own population data and the sorts of questions that they want to ask Yawuru people in order to develop their community according to the priorities of the community. So there's a different scenario.
So as in a conference we ended up with six of these local data sovereignty efforts and each one looks a bit different but each one of them raised the problems that we are now addressing. So would there be a potential for what we are doing to meet the needs of local communities? I think in many different ways the answer is yes so amongst the orphan datasets throughout the nation we suspect that land councils, Aboriginal medical services, Aboriginal legal services, [unclear] 56:31 corporations and other types of Aboriginal organisations have orphan datasets from years of operations that are not digitised and are not able to be used by the communities.
Now of course it’s impossible for us to say at this stage that an ideal world would be the digitisation of all of those datasets but we first of all need a survey of those datasets then to identify the ones with the highest priorities. I imagine that the highest priority would be at this stage the ones that are most vulnerable and at risk and also those datasets such as population datasets that are important for say critical health interventions.
C: The elves have told me that there are no Facebook questions yet. Please don’t let Facebook just be the home of polarised arguments. Let’s use it to pose some naughty but really important questions. I’m going to throw to the room for any – who’s going to ask the next question in the room? Oh there’s a hand up the back. Also another shy person but –
A: I’m Fiona Blackburn, I’m from AIATSIS. I’m interested, Len, in what you were saying about the orphan – the health surveys. In the context of the very local focus that Marcia’s been talking about some of those health surveys I should imagine would be national, cover great [unclear] 58:18 the relevant data might be returned to communities, to local communities that could use those subsets of that data.
L: Thanks Fiona. Well we’re entering a new era in Australia in terms of data access as a result of some Commonwealth initiatives essentially started by the Productivity Commission and the – what’s come out of that process is that as far as Commonwealth data is concerned a lot of the restrictions on access that existed in the past look like they’re going to be removed. Now that doesn’t mean that individual data is going to be made available to anybody who wants it but it means that we can see through the use of appropriate technology the sort of technology that’s commonplace in like law enforcement and national security or in the health sector, for that matter, where you have graded access to data depending on what your credentials are.
Now there’s a whole lot of specific issues involved there when you’re talking about indigenous data because of the issue of indigenous data sovereignty and ownership but the technology – we believe that the technology is there that will allow graded access to things like health surveys so that communities can get health – relationships with their doctors, that researchers can do epidemiological research drawing on the data that begins as individual health records but without jeopardising the confidentiality of those records. There’s a whole area of technology associated with what they call the semantic web which enables this to be done but we’re talking about something that is going to be some years in the development, I’m sure. But that’s what I was saying when I was talking, that this needs to be done as a partnership between really sophisticated computer science approaches and the communities.
C: Thank you for that. Another question here around the pillar.
A: Thanks, everybody. Anna Carr. I’m interested in the extent to which questions or the Indigenous Data Network can be of service. So you’ve spoken about the capacity for local groups and local organisations to use a network to meet their own priorities and have their population surveys etc used in combination with national survey datasets. Is there a – the capacity for the Indigenous Data Network to isolate and show how and where individual communities can use national datasets for their own purposes?
J: That’s a great question. One of the aspects of this project that distinguishes it from others is its emphasis on geospatial attribute data. One of the most important things that distinguishes indigenous data, one of the most important attributes is its very particular geographic patterning so that the definition of indigenous person is fundamentally a temporal and a geographic definition. These are people from Australia who’ve been there a very long them. If you drill down to more specific subsets of that attribute you have specific communities with specific kinship relationships that tie them to those geographic areas and to each other over very long periods of time and those kinship links in turn tie them into adjacent communities.
When we’re talking about things like health data or justice system engagement data they all have a geospatial dimension to them which is extremely informative and valuable to the community in terms of improving policy and service delivery based on that policy to that community. So knowing for example which members of which communities are accessing which health services in which areas. So how can government say – if it is government that’s providing a particular health service to a particular area – better adjust or better tailor the location of that service and the particular expertise that forms part of that service to that community just on the geographic attributes of the data that’s being collected. We have very, very powerful modelling and analytical techniques for helping government and other service providers to do that. The challenge is really nationalising it, making it national and standardising it nationally so that communities can have a coherent and cooperative involvement in the way that data is used for their purposes. Does it answer the question?
C: You look like you’re ready to respond, Len. Okay. Terrific. Now I should just say that the livestream will end at 11:15. We started late so we might just push on a bit beyond that. Now the livestream though will be available on the AIATSIS, NLA and Aus Preserves YouTube pages after the event so you will be able to revisit the event. I’m going to take a couple more questions if there are any in the room. We’ll have an opportunity for informal interaction once the event is finished but there is a question over here and over here. Well we’ll go here and then we’ll go here. You, sir.
A: Oh hi, I'm [unclear] 4:32, the National Mapping Agency and just to go on with that idea about geospatial data and the idea that it connects – like at GA we say that everything happens somewhere and we’ve started talking much more about digital data rather than geo – digital mapping, sorry, rather than geospatial data or location information just because people are familiar with it. But I’m also – I’m interested to know what – how place is being connected in or considered in the indigenous data network. We talk about the – having standards, national standards and being able to nationalise this. What specifically could we do in government to improve the way these things work?
L: Well James will probably want to say more but one of the things that would be great for us is if we had an indigenous geography which had a historical dimension as well, locations that are important in terms of indigenous history in Australia, no longer part of the standard gazetteer. We’re talking about names, the naming of geographical entities so that would be my response to that but there’s a lot of contemporary issues too and I think James is probably better to talk about that.
J: Look, it’s a phenomenally complicated question. With my native title hat on all of the data that we collected as part of the evidence development process for native title claims was fundamentally geographic so we developed a technique which you can read about online so there are videos where I’ve talked about this previously, in other forums, called event modelling for a population. So first of all you need to start – in order to really think about this in geographic terms start thinking about a population as a three-dimensional object made up – a single community, a single entity if you like constructed, not life expectancy and education level and so forth but characterised by links between them. Each link between a pair of individuals has a geographic expression. It’s a geographic expression that can be defined in a number of different ways.
If you take one geographic expression of a relationship between two individuals it’s the birth locations of those two individuals so straight away you can begin to unravel what I’m talking about when I mean that indigenous identity is a geographic identity because the birth locations of everybody in that single community is inside Australia for 2,000 generations at least. Then within that relational population model you have specific concentrations of relationships which are distinct to particular peoples with distinct languages and cultures and so forth. So for example we talk about Ngunawal people, these are people whose relationships recur over and over again in a particular geographic domain which is the region that we’re sitting within today.
Now we move there in that characterisation away from a strictly geospatial description to a temporal or temporospatial if you like expression so an event model is a model where each individual event brings together a time, a play to other events that includes another time and a place and an individual whether by kinship, whether by a productive relationship such as participation in a particular industry, whether by ceremonial participation, whether by use of the same language. You see where I’m going with this? Okay. So we begin to develop much more sophisticated, much more analytically powerful ways of understanding population dynamics not just for indigenous people but for all people but specifically for indigenous people where the question really is how do services controlled by indigenous communities get properly delivered? Tailored and delivered to specific communities as Craig pointed out earlier with different needs and as Marcia referred to earlier. I mean the needs and interests of communities in the East Kimberley are very different from those in La Perouse or in the western dessert where I grew up, for example.
But this requires a demography in health delivery, I mean this is a big endeavour so we’re starting at the bottom. The first thing to do is to recognise that we have a huge amount of data that’s been collected over more than 150 years which as Len pointed out sitting down in a mixed medium paper, some of it’s on microfilm, some of it’s been digitised, some of it hasn’t. The vast majority has not been indexed and this is valuable, this belongs to people and it has a monetary value not just in productivity dividends in terms of improved service delivery, productivity efficiency in terms of policy development but in terms of the property of real people just like land, just like labour. So as Marcia pointed out at the beginning of this Q and A session the first thing we need to do, and keeping in mind that orphan datasets are the priority, is to begin a nationwide survey, an audit to find out what actually is out there that is not known to the owners of that data so that we begin, just like we did after world war 2, begin to link back the stolen assets to the people that own it. Then begin to turn it into a more effective modelling and analytical device for those communities.
C: Thanks for the question, thanks for that. I was going to ask [unclear] 10:53 diaspora amongst indigenous Australians so I’m Djangadi here but I’m on Ngunawal land, I’m not in Kempsey. We have Djangadi in Melbourne, we have Djangadi in Adelaide so I have slight anxiety about over-geographicalising how we understand indigenous Australia because that may well obviate and render lots of us invisible ‘cause we’re not geographically defined anymore. There was a question over here. I’m told we can go a little bit over at 11:15 with the live feed for those online.
L: Could I just say some –
C: Marcia, yes.
L: Could I say something about your concerns about over-geographicalising the data? The kind of data that James was talking about motivated him to develop the approach that he has was actually genealogical data so James says that every entity has a geographical aspect they also have a genealogical aspect. So I wish that James had one of his gorgeous computer-generated images of what a deep genealogical dataset over several generations taking into account all the places where the people have resided and how they’ve intermarried. So that could show you how exactly what you’re talking about, how a core population and its diaspora can be captured by data and the attributes of that population are genealogical, geographical, historical and of course there are several other key attributes.
C: Super. While he’s here I’m going to have a look at one of his pictures then. Thanks Marcia. There’s a question over here and we have a question online so we’ll just go over here in the room first of all.
A: It may be the same question that is online because I posted –
C: Are you Mariam McBride?
C: Then it’s not.
A: I tried poting – I’m Katherine Styles, I work for the Registrar of Indigenous Corporations of which there are 3,000 around Australia. I’m just wondering whether you’re wanting people to – organisations to opt into this like in that survey, is it open access? Can we start doing it now? It’s very exciting.
: Just make contact.
A: Okay, thank you.
C: There is a website.
: And a Twitter account.
C: Well it’s really real now if it has a Twitter account. So does that answer your question?
A: Thank you. But James, why did you not show us one of these pictures?
C: Yeah, why, James? Why?
J: Yeah, why? I didn’t realise I was back in court, sorry, Your Honour. So I didn’t come here intending to present evidence, I came here intending to explain an approach but we can. I mean if the technicians are around we can unplug one computer and plug in another one if –
C: Let’s save that up, shall we? So I have a question from Miriam McBride which is a similar question, how do members of the public support this program? What can a regular citizen do to help? Who wants to have a crack at that? Marcia? Do you want to have a crack at that question?
L: Well I’m just grateful actually from what – I can’t see the audience but I can hear some people apprehending what we’re talking about. The more people that understand the urgent priority for an approach like this the more able we will be to convince governments and indeed communities to not be afraid of data, to share data and to assist the – I mean there’s lots of institutional support for this approach that would be most welcome. I think – those of us who have the orphan datasets of our own such as mine, it’s not quite orphaned yet but it results from three Australian Research Council linkage grants over roughly a 10-year period gathering together data on agreements within indigenous – it’s all online at ATNS.net.au. It’s had no ARC funding for a long time, it’s unlikely to get any more ARC funding and I’ve managed to keep it up-to-date with interns and scrabbling together 15,000 a year to keep it online. But if I can’t do that in the future it will be an orphaned dataset.
So we have to build into our thinking how do we prepare for datasets to be archived and yet remain accessible before they become orphans? So I’ve volunteered the dataset, the entire collection to – as a subject. as a case study to ensure that we get our approach right. So there’ll be a number of steps and protocols that we’ll have to develop and I’m sure that there are many other datasets like this hidden in universities and in institutions. Just coughing up the information to us will be an enormous help in the first instance.
C: Well thank you for that, Marcia, I’ve had – the elves have told me to wrap up so I am going to draw the formal part of our event to a close. But there will be some opportunity perhaps to engage the two speakers that are here at least in some informal conversation outside in the foyer so I invite you to do that. Thank you all very much for your participation in the event today. Can I just give Aus Preserves a bit of a plug? Communities of practice like this provide opportunities for practitioners, researchers to come together and to share their experience and perhaps save themselves from a bit of agony of starting from scratch when they don’t perhaps need to start from scratch ‘cause somebody’s already done something. So it allows a best practice corpus to be built. So I hope you’ve found this forum to be valuable and encouraging. Can I ask you to give a final round of applause to Professor Langton and Drs Smith and Rose for the presentations that they’ve given this morning and the information? My head’s swimming a little bit but would you thank them with a round of applause finally, please.
C: Look finally, I did notice there was a table in the Library bookshop with Marcia’s two books on it. Alas she’ll not be able to sign them because she’s not here. Geography’s important. So – but I think they’re available for sale in the bookshop and I would encourage you to avail yourself of those, thank you very much.
L: My apologies.
C: Thanks Mar –
End of recording