‘A map and some pins’: open data and unlimited horizons

8681035823_72f8c57ae0_m

This is the text of my keynote address to the Digisam conference on Open Heritage Data in the Nordic Region held in Malmö on 25 April 2013. You can also view the video and slides of my talk, or experience the full interactive experience by playing around with my text/presentation in all its LOD-powered glory. (Use a decent browser.)


The Australian poet and writer Edwin James Brady and his family lived for many years in the isolation of far eastern Victoria — in a little town called Mallacoota.

Edwin James Brady

Edwin James Brady (NLA: nla.pic-vn3704359)

Here, from time to time, Brady amused himself by taking a map of Australia down from the wall and sticking pins in it. The pins, Brady explained in 1938, included labels such as ‘Hydro-electric supply base’, ‘Irrigation area, and ‘Area for tropical settlement’. The map and its pins were one expression of Brady’s life-long obsession with Australia’s potential for development — for progress.

Maps and pins are probably more familiar now than they were in Brady’s time. We use them routinely for sharing our location, for plotting our travel, for finding the nearest restaurant. Maps and pins are one way that we document, understand and express our relationship to space.

Brady, however, was interested in using his pins to highlight possibilities. In the late nineteenth and early twentieth centuries size mattered. With the nations of Europe jostling for land and colonial possessions, space become an index of power. When the Australian colonies came together in 1901 to form a nation, maps and spatial calculations abounded. Australia was big and so it’s future was filled with promise.

Australia was big with promise.

Australia was big with promise.

In his travels around Australia, EJ Brady started to catalogue ways in which its vast, ‘empty’ spaces might be turned to productive use. A hardy yeomanry armed with the latest science could transform these so-called ‘wastes’, and Brady was determined to bring these opportunities to the attention of the world.

This evangelical crusade reached its high point in 1918, with the publication of his huge compendium, Australia Unlimited — 1139 pages of ‘Romance History Facts & Figures’.

National Archives of Australia:  A659, 1943/1/3907, page 135

National Archives of Australia:
A659, 1943/1/3907, page 135

Space may no longer be invested with the same sense of swelling power, but our maps and pins still figure in calculations of progress. Now it is the data itself that awaits exploitation. Our carefully plotted movements, preferences and passions may well have value to governments, planners or advertisers. Data, according to some financial pundits, ‘is the new oil’.

Whereas Brady traveled the land documenting its untapped riches, we can take his work and life and mine it for data — looking for new patterns and possibilities for analysis.

Brady wasn’t the first to use the phrase ‘Australia Unlimited’, though he did much to make it familiar. By exploring the huge collection of digitised newspapers available through the National Library of Australia’s discovery service, Trove, we can track the phrase through time.

australia_unlimited

Brady was a skilled self-publicist and his research trips in 1912 were eagerly reported by the local press such as the Barrier Miner in Broken Hill and the Cairns Post.

In 1918 the book was published, receiving a generally positive reception — as an advertising leaflet showed, even King George V thought the book was ‘of special interest’. In 1926, a copy of the book was presented to a visiting delegation from Japan.

NAA: A659, 1943/1/3907, page 55

NAA: A659, 1943/1/3907, page 55

Over the years Brady sought to build on his modest successes, planning a variety of new editions and even a movie. But while his hopes were thwarted, the phrase itself lived on.

In 1938 and 1952, there are references to a radio program called ‘Australia Unlimited’, apparently featuring young musical artists. Also in 1952 came the news that the author of Australia Unlimited, EJ Brady, had died in Mallacoota at the age of 83.

Sydney Morning Herald, 23 July 1952

Sydney Morning Herald, 23 July 1952

Unfortunately copyright restrictions bring this analysis to a rather unnatural end in 1954. If we were able to follow it through until the present, we could see that from 1957 the phrase was used by a leading daily newspaper as the title of an annual advertising supplement detailing Australia’s possibilities for development. In 1958, it was adopted as a campaign slogan by Australia’s main Conservative party. In 1999 it was resurrected again by a major media company for a political and business symposium. Even now it provides the focus for a ‘branding’ initiative supported by the federal government.

Graphs like this are pretty easy to interpret, but of course we should always ask how they were made. In this case I simply harvested the number of matching search results from the Trove newspaper database for each year. The tool I built to do this has been through several different versions and is now a freely-accessible web application called QueryPic. Anyone can quickly and easily create this sort of analysis just by entering a few keywords.

QueryPic is one product of my experiments with the newspaper database over the last couple of years. I’ve also looked at changes to content of front pages, I’ve created a combination discovery interface and fridge poetry generator, and I’ve even built a simple game called Headline Roulette — which I’m told is strangely addictive.

All of these tools and experiments take advantage of Trove’s API. I think it’s important to note that the delivery of cultural heritage resources in a machine-readable form, whether through a custom API or as Linked Open Data, provides more than just improved access or possibilities for aggregation. It opens those resources to transformation. It empowers us to move beyond ‘discovery’ as a mode of interaction to analyse, extract, visualise and play.

Using freely available tools we can extract named entities from a text, we can look for topic clusters across a collection of documents, we can find places and pin them to a map. With a little bit of code I can take the newspaper reports of Brady’s travels in 1912 and map them. With a bit more time I could take another of Brady’s travel books, River Rovers, available in digitised form through the Internet Archive, and plot his journey along Australia’s longest river, the Murray.

Such transformations help us see resources in different ways — we can find new patterns, new problems, new questions. But transformation is a lossy business. I can put a pin in a map to show that Brady stopped off in Mildura on his voyage along the Murray. What is much harder to represent are the emotions that surrounded that visit. While he was there Brady received news of a friend’s death. ‘Bad news makes hateful the most pleasant place of abiding’, he wrote mournfully, ‘I strained to open the gate to go forth again into a wilderness of salt bush and sere sand’. Travel can be a form of escape.

Digital humanist and designer Johanna Drucker has written about the problems of representing the human experience of time and space using existing digital tools. ‘If I am anxious’, she notes, ‘spatial and temporal dimensions are distinctly different than when I am not’. We do not experience our journeys in a straightforward linear fashion — as the accumulation of metres and minutes. We invest each footstep with associations and meanings, with hopes for the future and memories of the past. Drucker calls on humanities scholars to articulate these complexities and work towards the development of new techniques for modelling and visualising our data.

‘If human beings matter, in their individual and collective existence, not as data points in the management of statistical information, but as persons living actual lives, then finding ways to represent them within the digital environment is important.’

In a similar way, Australia Unlimited is not just a catalogue of potentialities, or a passionate plea for national progress. It’s also the story of a struggling poet trying to find some way of supporting his family. Proceeds from the book enabled Brady to buy a plot of land in Mallacoota and build a modest home — I suspect that it wasn’t the same home that was painted by his wife in the 1950s.

nla.pic-an2287718-v

But even this small success was undermined. Distribution of the book was beset with difficulties and disappointments and and despite all his plans financial security remained elusive.

Brady’s youngest daughter, Edna June, was born to his third wife Florence in 1946. What could he leave he leave her? A ‘modern edition’ of Australia Unlimited lay completed but unpublished. ‘If it fails to find a publisher’, he remarked wistfully, ‘the MSS will be a liberal education for her after she has outgrown her father’s nonsense rhymes’. It was, he pondered, ‘a sort of heritage’.

One of the things I love about being a historian is that the more we focus in on the past the more complicated it gets. People don’t always do what we expect them to, and that’s both infuriating and wonderful.

Likewise, while we often have to clean up or ‘normalise’ cultural heritage data in order to do things with it, we should value its intrinsic messiness as a reminder that it is shot through with history. Invested with the complexities of human experience it resists our attempts at reduction, and that too is both infuriating and wonderful.

The glories of messiness challenge the extractive metaphors that often characterise our use of digital data. We’re not merely digging or mining or drilling for oil, because each journey into the data offers new possibilities — our horizons are opened, because our categories refuse to be closed. These are journeys of enrichment, interpretation and creation, not extraction.

We’re putting stuff back, not taking it out.

Cultural institutions have an exciting opportunity to help us work with this messiness. The challenge is not just to pump out data, anyone can do that. The challenge is to enrich the contexts within which we meet this data — to help us embrace nuance and uncertainty; to prevent us from taking the categories we use for granted.

For all it’s exuberant optimism, a current of fear ran through Australia Unlimited. The publisher’s prospectus boldly announced that it was a ‘Book with a Mission’. ‘A mere handful of White People’, perched uncomfortably near Asia’s ‘teeming centres of population’, could not expect to maintain unchallenged ownership of the continent and its potential riches. Australia’s survival as a white nation depended upon ‘Effective Occupation’, secured by a dramatic increase in population and the development of its vast, empty lands — ‘The Hour of Action is Now!’.

National Archives of Australia:  A659, 1943/1/3907, page 208

National Archives of Australia:
A659, 1943/1/3907, page 208

In 1901, one of the first acts of the newly-established nation of Australia was to introduce legislation designed to keep the country ‘white’. Restrictions on immigration, administered through a complex bureaucratic system, formed the basis of what became known as the White Australia Policy.

While the legislation was designed to keep non-white people out, an increase of the white population was seen as essential to strengthen security and legitimise Australia’s claim to the continent. Australia Unlimited was an exercise in national advertising aimed at filling the unsettling emptiness with sturdy, white settlers.

But White Australia always a myth. As well as the indigenous population there were, in 1901, many thousands of people classified as non-white living in Australia. They came from China, India, Indonesia, Turkey and elsewhere. A growing number had been born in Australia. They were building lives, making families and contributing to the community.

Here are some of them…

The real face of White Australia

The real face of White Australia

I built this wall of faces using records held by the National Archives of Australia. If a non-white person resident in Australia wanted to travel overseas they needed to carry special documents. Without them they could be prevented from re-entering the country — from coming home. Many, many thousands of these documents are preserved within the National Archives.

Kate Bagnall, a historian of Chinese-Australia, and I are exploring ways of exposing these records through an online project called Invisible Australians.

To build the wall I downloaded about 12,000 images from the Archives’ website — representing just a fraction of one series relating to the administration of the White Australia Policy. Unfortunately there’s no machine-readable access to this data, so I had to resort to cruder means — reverse-engineering interfaces and screen-scraping.

Once I had the images I ran them through a facial detection script to find and crop out the portraits. What we ended up with was a different way of accessing those records — an interface that brings the people to the front; an interface which is compelling, discomfiting, and often moving.

The wall of faces also raises interesting questions about context. Some people might be concerned by the loss of context when images are presented in this way, although each portrait is linked back to the document it was derived from, and to the Archive’s own collection database. What is more important, I think, are the contexts that are gained.

If you’re viewing digitised files on the National Archives’ own website, you can only do so one page at a time. Each document is separate and isolated. What changes when you can see the whole of the file at once? I’ve built another tool that lets you do just that with any digitised file in the Archives’ collection. You see the whole as well as the parts. You have a sense of the physical and conceptual shape of the file.

National Archives of Australia:  ST84/1, 1908/471-480

National Archives of Australia:
ST84/1, 1908/471-480

In the case of the wall of faces, bringing the images together, from across files, helps us understand the scale of the White Australia Policy and how it impacted on the lives of individuals and communities. These were not isolated cases, these were thousands of ordinary people caught up in the workings of a vast bureaucratic system. The shift of context wrought by these digital manipulations allows us to see, and to feel, something quite different.

And we can go the other way. In another experiment I created a userscript to insert faces back into Archives’ website. A userscript is just a little bit of code that rewrites web pages as they load in your browser. In this case the script grabs images relating to the files that you’re looking at from Invisible Australians.

So instead of this typical view of search results.

Before

Before

You see something quite different.

After

After

Instead of just the record metadata for an individual item, you see that there are people inside.

We also have to remember that the original context of these records was the administration of a system of racial surveillance and exclusion. The Archives preserves not only the records, but the recordkeeping systems that were used to monitor people’s movements. The remnants of that power adhere to the descriptive framework. There is power in the definition of categories and the elaboration of significance.

Thinking about this I came across Wendy Duff and Verne Harris’s call to develop ‘liberatory standards’ for archival description. Standards, like categories, are useful. They enable us to share information and integrate systems. But standards also embody power. How can we take advantage of the cooperative utility of standards while remaining attuned to the workings of power?

A liberatory descriptive standard, Duff and Harris argue: ‘would posit the record as always in the process of being made, the record opening out of the future. Such a standard would not seek to affirm the keeping of something already made. It would seek to affirm a process of open-ended making and re-making’.

‘Holes would be created to allow the power to pour out.’

‘Making and re-making’ — sounds a lot like the open data credo of ‘re-use and re-mix’ doesn’t it? I think it’s important to carry these sorts of discussions about power over into the broader realm of open data. After all, open data must always, to some extent, be closed. Categories have been determined, data has been normalised, decisions made about what is significant and why. There is power embedded in every CSV file, arguments in every API.

This is inevitable. There is no neutral position. All we can do is encourage re-use of the data, recognising that every such use represents an opening out into new contexts and meanings. Beyond questions of access or format, data starts to become open through its use. In Duff and Harris’s words, we should see open data ‘as always in the process of being made’.

What this means for cultural institutions is that the sharing of open data is not just about letting people create new apps or interfaces. It’s about letting people create new meanings. We should be encouraging them to use our APIs and LOD to poke holes in our assumptions to let the power pour out.

There’s no magic formula for this beyond, perhaps, building confidence and creating opportunities. But I do think that Linked Open Data offers interesting possibilities as a framework for collaboration and contestation — for making and challenging meanings.

We tend to think about Linked Open Data as a way of publishing — of pushing our data out. But in fact the production and consumption of Linked Open Data are closely entwined. The links in our data come from re-using identifiers and vocabularies that others have developed. The linked data cloud grows through a process of give and take, by many small acts of creation and consumption.

There’s no reason why that process should be confined to cultural institutions, government departments, business, or research organisations. Linked Open Data enables any individual to talk about what’s important to them, while embedding their thoughts, collections, passions or obsessions within a global conversation. By sharing identifiers and vocabularies we create a platform for communication. Anyone can join in.

So, if we want people to engage with our data, perhaps we need to encourage them to create their own.

I’ve just been working on a project with the Mosman public library in Sydney aimed at gathering information about the experiences of local servicepeople during World War One. There are many such projects happening around the world at the moment, but I think ours is interesting in a couple of ways. The first is a strong emphasis on linking things up.

The are records relating to Australian service people available through the National Archives, the Australian War Memorial, and the Commonwealth War Graves Commission, but there are currently no links between these databases. I’ve created a series of screen scrapers that allow structured data to be easily extracted from these sources. That means that people can, for the first time, search across all these databases in one hit. It’s a very simple tool that I started coding to ease the boredom of a long bus trip — but it has proved remarkably popular with family historians.

Once you’ve found entries in these databases, you can just cut and paste the URL into a form on the Mosman website and a script will retrieve the relevant data and attach it to the record of the the person you’re interested in. Linking to a service record in the National Archives, for example, will automatically create entries for the person’s birthplace and next-of-kin.

The process of linking builds structures, and these structures will themselves all be available as Linked Open Data. Even more exciting is that the links will not only be between the holdings of cultural institutions. The stories, memories, photographs and documents that people contribute will also be connected, providing personal annotations on the official record.

None of this is particularly hard, it’s just about getting the basics right. Remembering that structure matters and that links can have meaning. It’s also about recognising that ‘crowd sourcing’ or user-generated content can be made anywhere. Using Linked Open Data people can attach meanings to your stuff without visiting your website. Through the process of give and take, creation and consumption, we can build layers of description, elaboration, and significance across the web.

What excites me most about open cultural data is not the possibility of shiny new apps or collection visualisations, but the possibility of doing old things better. The possibility of reimagining the humble footnote, for example, as a re-usable container of structured contextual data — as a form of distributed collection description. The possibility of developing new forms of publication that immerse text and narrative within a rich contextual soup brewed from the holdings of cultural institutions.

I want every online book or article to be a portal. I want every blog or social media site to be a collection interface.

What might this look like? Perhaps something like this presentation.

My slides today are embedded within a HTML document that incorporates all sorts of extra goodies. The full text of my talk is here, and as you scroll through it you’ll see more information about the people, places and resources I mention pop up in the sidebar. Alternatively you can explore just the people or the resources, looking at the connections between them and the contexts in which they’re mentioned within my text.

This is part of an ongoing interest to explore different ways of presenting historical narrative, that build a relationship between the text and the structured data that underlies the story.

All of the structured data is available in machine-readable form as RDF — it is, in itself, a source of LOD. In fact the interface is built using an assortment of JavaScript libraries that read the RDF into a little temporary triplestore, and then query it to create the different views. So the whole thing is really powered by LOD.

It’s still very much an experiment, but I think it raises some interesting possibilities for thinking about how we might consume and create LOD simply by doing what we’ve always done — telling stories.

EJ Brady’s dreams were never realised. Australia’s vast spaces remained largely empty, and the poet continued to wrestle with personal and financial disappointment. ‘After nearly eight decades association with the man’, Brady wrote of himself in 1949, ‘I have come to look upon him as the most successful failure in literary history’. This energetic booster of Australia’s potentialities was well aware of his own life’s mocking irony. ‘He has not… made the wages of a wharf laborer out of book writing yet he persists in asserting Australia is the best country in the world!’.

But still Brady continued to add pins to his map.’For half a century I’ve been heaping up notes, reports, clippings, pamphlets, etc. on… all phases of the country’s life and development’. ‘What in hell I accumulate such stuff for I don’t know’, he complained in 1947. As the elderly man surveyed the ‘bomb blasted pile of rubbish’ strewn about his writing tent, he admitted that ‘this collecting is a sort of mania’.

Brady’s map and pins told a complex story of hope and disappointment, of confidence and fear. A story that combined national progress with an individual’s attempts merely to live.

There are stories in our data too — complex and contradictory stories full of emotion and drama, disappointment and achievement, violence and love. Let’s find ways to tell those stories.

Exposing the archives of White Australia

I recently gave a presentation in the Institute of Historical Research’s Digital History Seminar series. The time difference between London and Canberra was a bit of a challenge, so I pre-recorded the presentation and then sat in my own Twitter backchannel while it played. For the full podcast information go to HistorySPOT. You can also play with my slides or peruse the #dhist Twitter archive.

Exposing the Archives of White Australia from History SPOT on Vimeo.

Too important not to try

On Friday 19 October I joined an enthusiastic group of digital humanities explorers at a Deakin University event entitled Dipping a Toe into the Digital Humanities and Creative Arts. @catspyjamasnz has assembled an excellent summary of the day in Storify.

In the morning I told the story of Invisible Australians. You can view the slides of Too or important not to try and listen to my dodgy audio recording via SoundCloud.

In the afternoon I gave a whirlwind workshop which included a headline roulette smackdown and an introduction to the wonders of Zotero.

Topic modelling in the archives

There seems to be a lot of topic modelling going on at the moment. Any why not? Projects like Mining the Dispatch are demonstrating the possibilities. Tools like Mallet are making it easy. And generous DHers like Ted Underwood and Scott Weingart are doing a great job explaining what it is and how it works.

I’ve talked briefly about using topic modelling to explore digitised newspapers, something that the Mapping Texts project has also been investigating. But I’ve also been following with interest Chad Black’s use of algorithmic techniques, including topic modelling, to look for local variations amidst the legal system of the early modern Spanish empire.

As part of the Invisible Australians project, Kate and I are exploring the bureaucracy of the White Australia Policy. In particular, we’re interested in the interaction between policy and practice, between the highly-centralised bureaucracy and the activities of individual port officials. Like Chad, we’re interested in mapping local variations — to try and understand the bureaucracy from the point of view of an individual forced to live within its restrictions.

I recently gave a presentation about the project at Digital Humanities Australasia (post coming soon!), and in preparation I decided to try a few topic modelling experiments. They were very simple, but I was impressed by the possibilities for exploring archival systems.

The problem I started with was this. The workings of the White Australia Policy are well documented by records held by the National Archives of Australia. Some series within the archives are specifically related to the operations of the policy — such as those containing many thousands of CEDTs. But there are also general correspondence series created by the customs offices in each state, as well as the Commonwealth Department of External Affairs which administered the Immigration Restriction Act (responsibility was later taken by the Department of Home and Territories and it’s successors). These general correspondence series are important, because they often include details of difficult or controversial cases — those that required a policy judgment, or prompted a change in existing practices. But how do you find relevant files within series that can contain large numbers of items?

Series A1, for example, is a correspondence series created by the Department of External Affairs. It contains more than 60,000 items. Past research tells us that amongst these 60,000 files are records of important policy discussions relating to White Australia. But these files tend to be labelled with the names of the people involved, so unless you know the names in advance they can be difficult to find.

Mitchell Whitelaw’s A1 Explorer, part of the Visible Archive project, lets you to explore the contents of Series A1 in a easy and engaging way. But while the A1 Explorer provides new opportunities for discovery, it doesn’t offer the fine-grained analysis we need to sift out the files we’re after. And so… topic modelling.

The process was pretty simple. While I can dip into my bag of screen-scrapers to harvest series directly from the NAA’s RecordSearch database, there was already an XML dump of A1 available from data.gov.au. So I extracted the basic file metadata from the XML and wrote the identifiers and titles out to a text file, one item per line. Following the instructions on the website I then loaded this file into Mallet:

/Applications/Mallet/bin/mallet import-file --input ./A1.txt --output A1.mallet --keep-sequence --remove-stopwords

Then it was just a matter of firing up the topic modeller:

/Applications/Mallet/bin/mallet train-topics --input ./A1.mallet --output-state ./A1.gz --output-doc-topics ./A1-topics.txt --output-topic-keys ./A1-keys.txt --num-topics 40

Again, I just followed the examples on the Mallet site.

Once it was finished I opened up A1-keys.txt to browse the ‘topics’ Mallet had found. The results were intriguing. There are a large number of applications for naturalisation in A1, so it’s no surprise that ‘naturalisation’ figures prominently in a number of the topics. What was more interesting was the way Mallet had grouped the naturalisation files. For example:

naturalization christian hans hansen jensen petersen andersen nielsen larsen christensen johannes jens niels pedersen andreas johansen martin jorgensen

and

naturalisation certificate giuseppe salvatore frank la leo samios spina sorbello leonardo fisher natale patane torrisi barbagallo luka rossi ross

Based on the co-occurrence of names within the file titles, Mallet had created groupings that roughly reflected the ethnic origins of applicants. It makes sense when you think about what Mallet is doing, but I still found it pretty amazing.

Mallet also found clusters around the major activities of the department, such as the administration of the territories. But of most interest to us was:

1 0.55539 passport ah student exemption students lee wong chinese young deserter education sing wing chong readmission son hing chin wife

The Chinese names alongside words such as ‘readmission’ and ‘wife’ suggested that this topic revolved around the administration of the White Australia Policy. This was easy to test. In A1-topics.txt was a list of every file in the series and their weightings in relation to each of the topics. I wasn’t sure what was a reasonable cut-off value to use in assessing the weightings, but after a bit of trial and error I fixed on a value of 0.7. I then just extracted the identifiers of every file that had a weighting greater than 0.7 for this topic. I used the identifiers to build a simple web page that Kate and I could browse. I also included links back to RecordSearch so we could explore further.

Browse the full list

It’s a pretty impressive result. Instead of fumbling with the uncertainties of keyword searches, we now have a list of more than 1,300 files that are clearly of relevance to Invisible Australians. There’s a few false positives and there are likely to be other files that we’ll have missed altogether, but now we have a much clearer picture of the types of files that are included and how they are described.

And that was at my first attempt, simply using the default settings. I’m now starting to play around with some of Mallet’s configuration options to see what sort of difference they make. I’m also keen to try out GenSim, a topic modelling package for Python.

I’m really excited about the possibilities of these sort of tools for analysing the contents of archival descriptive systems, something I mentioned in my Digital Humanities Australasia paper. Much more to come on this I suspect…

It’s all about the stuff: collections, interfaces, power and people

This is the full version of a paper I presented at the National Digital Forum, 30 November 2011.

In 1901, one of the first acts of the Commonwealth of Australia was to create a system of exclusion and control designed to keep the newly-formed nation ‘white’. But White Australia was always a myth. As well as the Indigenous population, there were already many thousands of people classified as ‘non-white‘ living in Australia — most were Chinese, but there were also Japanese, Indians, Syrians and Indonesians.

Here are some of them…

The real face of White Australia

The administration of what became known as the White Australia Policy created a huge volume of records, much of which is still preserved within the National Archives of Australia. These photographs are attached to certificates that non-white residents needed to get back into the country if they decided to travel overseas. There are thousands upon thousands of these certificates in the Archives. Thousands of certificates representing thousands of lives — all monitored and controlled.

But is is too easy to see these people as the powerless victims of a repressive system. There were many acts of resistance. Some argued against the need to be identified ‘just like a criminal’. Others exercised control over their representation, submitting formal studio portraits instead of mug shots.

Most commonly and most powerfully, people resisted the policy simply by going ahead and living rich and productive lives.

My partner, Kate Bagnall, is helping to rewrite Australian-Chinese history by overthrowing the stereotype of the culturally isolated Chinese man living a lonely, meagre existence surrounded by gambling and opium dens. By mining the available records, by reading against the grain of contemporary reports and by working with family historians, Kate is documenting their intimate lives — their wives, their lovers, their families and descendants — the sorts of relationships that sent a shudder through the edifice of White Australia. Power can be reclaimed in many subtle and subversive ways.

‘The real face of White Australia’ is an experiment. It uses facial detection to technology to find and extract the photographs from digital copies of the original certificates made available through the National Archives of Australia’s collection database. The photographs you see here come from just one series, ST84/1. There’s no API to the collection so I reverse-engineered the web interface to create a script that would harvest the item metadata and download copies of all the digitised images. There are 2,756 files in this series. On the day I harvested the metadata, 347 of those files had been digitised, comprising 12,502 images. It took a few hours, but I just ran my script and soon I had a copy of all of this in my local database.

Then came the exciting part. Using a facial detection script I found through Google and an open source computer vision library, I started experimenting with ways of extracting the photos. After a few tweaks I had something that worked pretty well, so I pointed my aging laptop at the 12,502 images and watched anxiously as the CPU temperature rose and rose. It took a few emergency cooling measures, but the laptop survived and I had a folder containing 11,170 cropped images. About a third of these weren’t actually faces, but it was easy to manually remove the false positives, leaving 7,247 photos.

These photos. These people.

With my database fully primed and loaded it was just a matter of creating a simple web interface using Django for the backend and Isotope (a jQuery plugin) at the front. Both are open source projects. All together, from idea to interface, it took a bit more than a weekend to create, and most of that was waiting for the harvesting and facial detection scripts to complete. It would be silly to say it was easy, but I would say that it wasn’t hard.

What we ended up with was a new way of seeing and understanding the records — not as the remnants of bureaucratic processes, but as windows onto the lives of people. All the faces are linked to copies of the original certificates and back to the collection database of the National Archives. So this is also a finding aid. A finding aid that brings the people to the front.

According to Margaret Hedstrom the archival interface ‘is a site where power is negotiated and exercised’. Whether in a reading room or online, finding aids or collection databases are ‘neither neutral nor transparent’, but the product of ‘conscious design decisions’. We would like to think that this interface gives some power back to the people within the records. Their photographs challenge us to do something, to think something, to feel something. We cannot escape their discomfiting gaze.

But this interface represents another subtle shift in power. We could create it without any explicit assistance or involvement by the National Archives itself. Simply by putting part of the collection online, they provided us with the opportunity to develop a resource that both extends and critiques the existing collection database. Interfaces to cultural heritage collections are no longer controlled solely by cultural heritage institutions.

It’s these two aspects of the power of interfaces that I want to focus on today.

There are a growing number of examples where the records created by repressive or discriminatory regimes have, in Eric Ketelaar’s words, ‘become instruments of empowerment and liberation, salvation and freedom’. Nazi records of assets confiscated during the Holocaust have been used to inform processes of restitution and reparation. Government records have helped members of Australia’s Stolen Generations trace family members. Descendants of inmates incarcerated by American colonial authorities in what was the world’s largest leprosy colony in the Philippines, have embraced the administrative record as an affirmation of their own heritage and survival. Records can find new meanings. Power can be reclaimed.

Technology can help. Tim Hitchcock has described how something as simple as keyword searching can turn archives on their heads. Recordkeeping systems tend to reflect the structures and power relations of the organisations that create them. The ‘hierarchical and institutional nature of most archives’, Hitchcock argues, ‘contains an ideological component which is sucked in with every dust-filled breath’. But digitisation and keyword searching free us from having to follow the well-worn paths of institutional power. We can find people and follow their lives against the flow of bureaucratic convenience. We can gain a wholly new perspective on the workings of society. ‘What changes’, Hitchcock asks, ‘when we examine the world through the collected fragments of knowledge that we can recover about a single person, reorganised as a biographical narrative, rather than as part of an archival system?’

Projects such as Unknown no longer may help us answer that question.

Unknown no longer

It’s aiming to extract the names and biographical details of slaves from the 8 million manuscript documents held by the Virginia Historical Society. The documents include court records, receipts, wills and inventories. Here is a page from the ‘Inventory of Negroes at Berry Plain Plantation, King George County, Virginia’ for 1855, listing names, occupations and valuations.

Tim Hitchcock is one of the directors of London Lives a project that similarly seeks to find the people in 240,000 manuscript pages documenting the lives of plebeian Londoners in the 17th century.

London Lives

More than three million names have already been extracted from the records of courts, workhouses, hospitals and other institutions. Work is continuing to link these names together, to merge these various shards of identity and piece together the experiences of London’s poorest inhabitants.

Remember me from the US Holocaust Memorial Museum is working with photographs taken by relief agencies in the aftermath of World War Two. The photographs are of displaced children who survived the Holocaust but were separated from families. What happened to them? The project is seeking public help to identify and trace the children.

Remember me

These are all projects about finding people. Finding the oppressed, the vulnerable, the displaced, the marginalized and the poor and giving them their place in history. This is what Kate and I hope to do with Invisible Australians, the broader project of which our faces experiment is part.

Invisible Australians

‘Invisible Australians’ aims to extract more than just photographs. We want to record and aggregate the biographical data contained within the records of the White Australia Policy — to extract the data and rebuild identities.

But we want to do more, we want to link these identities up with with other records, with the research of family and local historians, with cemetery registers and family trees, with newspaper articles and databases we don’t even know about yet. We want to find people, families and communities.

It’s ridiculously ambitious and totally unfunded. But it is possible.

The most exciting part of online technology is the power it gives to people to pursue their passions. As with the faces, we don’t need the help of the National Archives. We need the records to be digitized, but that’s happening anyway and we can afford to be patient. Most of the tools we need already exist, and are free. In the past 12 months, for example, there have been a number of open source tools released for crowd-sourced transcription of manuscript records.

People with passions, people with dreams, people who are just annoyed and impatient, don’t have to wait for cultural institutions to create exactly what they need. They can take what’s on offer and change it.

Interfaces can be modified. It is amazingly easy to write a script that will change the way a web page looks and behaves in your browser. I was frustrated by the standard interface to digitized files in the National Archives of Australia’s Recordsearch database — so I changed it.

Before and after

Not only did make it look a bit nicer, I added new functions. My script lets you print a whole file or a range of pages and display the entire contents of the file on a pretty cool 3d wall.

I’ve shared this script, and a few other Recordsearch enhancements. Anyone can install them with a click and use them.

Wragge Labs Emporium

Interfaces are sites of power and we can claim some of that power for ourselves. Online technologies not only free us from the having to brave the physical intimidation of the reading room, they free us up to engage with the records in new ways. The archivist-on-duty would probably not be pleased if I pulled out some scissors and started snipping photos out of certificates. Or if I pulled a file apart and pasted it’s contents on the wall. But online we are free to experiment.

The power of cultural heritage organisations is perhaps expressed most forcefully in their ability to control the arrangement and description of their collections. ‘Every representation, every model of description, is biased’, note Verne Harris and Wendy Duff, ‘because it reflects a particular world-view and is constructed to meet specific purposes’. Archives, libraries and museums are already starting to share this power, by allowing tagging, or seeking public assistance with description through crowd sourcing projects. But most of the these activities still happen within spaces created and curated by the institutions themselves. Our cathedrals of culture might be opening their doors and inviting the public to participate in their ceremonies, but that doesn’t make them bazaars. The architecture stills speaks of authority.

In any case, people already have a space where they can explore and enrich collections — it’s called the internet.

It would be great to see cultural institutions doing more to watch, understand and support what people are doing with collections in their own spaces — following them as they pursue their passions, rather than thinking of ways to motivate them.

A quick example… You might have heard of Zotero, it’s an open source project that lets you capture, annotate and organize your research materials.

Zotero

One cool thing about Zotero is that you can build and contribute little screen scrapers, called translators, that let Zotero extract structured data from any old collection database. You might not be surprised to learn that I’ve created a translator for Recordsearch. Another cool thing about Zotero is that you can share the stuff that you collect in public groups.

Invisible Australians Zotero group

Put those two cool things together and what do you have? Well to me they spell out user generated finding aids — parallel collection databases created by researchers simply pursuing their own passions.

Linked Open Data greatly increases opportunities for collection description to leak into the wider web. If objects and documents are identified with a unique URL, then anyone can can make and publish statements about them in machine-readable form. These statements can then be aggregated and explored. Initiatives such as the Open Annotation Collaboration will hasten the development of these shared descriptive and interpretative layers around our cultural collections.

And of course all this descriptive and interpretative work can be harvested back to enhance existing collection databases. We could start doing it now — though I will spare you today my rant about the possibilities of mining footnotes.

As well as exploring the possibilities of user-generated content, cultural institutions are starting to open up their collection data for re-use. APIs are great (though Linked Open Data is better), and New Zealand is lucky to have an organisation like DigitalNZ which just gets it. People can and will make cool things with your stuff.

But again, we don’t have to wait for everything to be delivered in a convenient, machine-readable form. If it’s on the web anybody can scrape, harvest and experiment.

You probably all know about the National Library of Australia’s newspaper digitisation project — it’s building a magnificent resource. But I wanted to do more than just find articles. I wanted to explore and analyze their content on a large scale. So I built a screen scraper to extract structured data from search results, and then used the scraper to  power a series of tools. I have a harvester that lets you download an entire results set — hundreds or thousands of articles — with metadata neatly packaged for further analysis.

Harvester

Or what about a script that graphs the occurrence of search terms over time, and allows you to ask questions like When did the Great War become the First World War?.

When did the Great War become the First World War?

In the end I got a bit carried away and built my own public API to the Trove newspaper database.

Unofficial Trove newspapers API

I think it’s important to note that the tools I developed were guided by the types of questions I wanted to ask. While we should welcome APIs and celebrate their possibilities, we should also remain critical. APIs are interfaces, they too embed power relations. Every API has an argument. What questions do they let us ask? What questions do they prevent us from asking?

Even as we move from the age of lumbering, slow-witted data silos into the rapidly-evolving realms of Linked Open Data, we have to constantly question the models we make of the world. Ontologies and vocabularies are culturally determined and historically specific. Yes, they too are interfaces, complete with their own distributions of power and authority. But we can revisit and change them. And we can relate our new models to our old models, capturing complex, long-term shifts in the way we think about the world. That’s incredibly exciting.

All of this hacking, harvesting, questioning, enriching and meaning-making makes me think about the possibilities of grassroots leadership. Online technologies enable people to take cultural institutions into unexpected realms. They can build their own interfaces, ask their own questions, determine their own needs — they can point the way instead of simply waiting to be served.

You might wonder what the National Library of Australia thinks of my various scrapers and harvesters. I can’t speak for them, but I can say that they’ve awarded me a fellowship to explore further the possibilities of text-mining in their newspaper database.

The idea of grassroots leadership brings me back to the title of this talk — ‘It’s all about the stuff’. It seems to me that we tend to model the interactions between cultural institutions and the public as transactions. The public are ‘clients’, ‘patrons’, ‘users’ or ‘visitors’. But the sorts of things I’ve been talking about today give us a chance to put the collections themselves squarely at the centre of our thoughts and actions. Instead of concentrating on the relationship between the institution and the public, we can can focus on the relationship we both have with the collections.

It’s all about the stuff.

It’s all about the respect and responsibility we both have for our collections.

It’s all about the respect and responsibility we both have for people like this.

 

 

Every story has a beginning

Entering the web of data

[view the presentation...] [view the triples...]

Keynote delivered at the annual conference of the Australia and New Zealand Society of Indexers, 14 September 2011.


This is me.

Today, Wednesday, 14 September 2011, I’m honoured to be able to join you here in the luxurious surrounds of the Brighton Savoy Hotel for the ‘Indexing See Change‘ conference. This is an event, a moment in history; we can pinpoint ourselves, this gathering, both in time and in space.

If we do that, if we move outside the moment and position ourselves on a timeline or a map, interesting things start to happen. Connections emerge.

Here we are at number 150, The Esplanade, in Brighton. A bit over a kilometre away is the stately villa, Kamesburgh. For many years Kamesburgh was also known as the Anzac Hostel — a refuge for permanently-incapacitated World War One veterans.

The Anzac Hostel opened on 5 July 1919. Here it is draped in its patriotic finery, from the collections of the Australian War Memorial. According to the caption, the Anzac Hostel was ‘a home, not an institute’.

Also amongst the War Memorial’s holdings is a wheeled bed that was used at the hostel. This particular bed was apparently occupied by one man, Albert Ward, for forty-three years.

Death notice for Alexander Kelley. Argus, 29 January 1944.

It was probably in a bed just like this that Alexander Dewar Kelley passed away on 27 January 1944. Alexander Kelley was cremated, and his remains interred amongst the roses at what is now called the Springvale Botanical Cemetery. Not far from my own grandparents.

Alexander Kelley spent close to half his life in the Anzac Hostel. Like many young men, he bravely answered his nation’s call to arms, but returned from war much changed. We can follow Alex’s war through his service record, easily-accessible through the website ‘Mapping Our Anzacs‘.

Alex was a coach painter who enlisted in the AIF in January 1916. Within a year he was in France. In May 1917 he suffered a gunshot wound to the head, but was able to rejoin his unit in August. Less than a month later though, he was wounded again, this time more severely. For Alex the war was over, and he was shipped back to Australia in May 1918.

‘Mapping Our Anzacs’ includes a scrapbook feature through which visitors to the site can attach notes or photographs to a service record. Amongst the the many thousands of postings is a fragment from a diary, found tucked inside the bible of Alexander Kelley’s mother. The diary entry reads simply: ‘Alex arrived from Front. Wet day. Saw him at “Caulfield”.’

Alex had survived and had returned to his family. This was a day to remember. But there was sadness too, for Alex was not the same young man who had left for the battlefields of Europe. In the diary fragment, ‘Caulfield’ is enclosed in inverted commas, indicating perhaps that the reunion took place, not in the suburb, but in the Caulfield rehabilitation hospital. Alexander Kelley was wounded in the face, hands and legs. He was left blind in both eyes and his right leg was amputated. He would live the remainder of his life a little over a kilometre away from here at the Anzac Hostel.

This is just one story. There are over 375,000 World War One service records held by the National Archives of Australia. How can we hope to understand a number like that? How can we hope to imagine the war’s impact on families, on communities?

‘Mapping Our Anzacs’ uses familiar Google maps to display the places of birth and enlistment recorded in many of those service records. But technical limitations make it impossible to display all the places at once. You can, however, take the same data and open it in Google Earth. If you then zoom in on Victoria, you see something like this.

Mapping Our Anzacs data viewed in Google Earth.

Each marker represents a place where a service person was born or enlisted. It’s impossible to read, of course, but that’s the point. There is so little blank space. As you zoom further, more markers appear, more place names resolve. It’s simple, but it’s powerful. They came from everywhere. From the smallest village to the biggest city; nowhere was untouched.

The ‘Mapping Our Anzacs’ scrapbook offers another perspective. It’s possible to extract the images posted to the scrapbook and present them on a 3D wall. Amidst an assortment of memorabilia, there are faces. Not places, or records — this is a wall of people.

Mapping Our Anzacs Scrapbook photos viewed through CoolIris

It’s worth noting too that like the markers on the maps, these faces link back to the actual service records. So they’re not just a new way of seeing the collection, they’re a new way of exploring it.

But the records don’t stand in isolation, they themselves have a context. A couple of years ago, Mitchell Whitelaw from the University of Canberra, undertook a project called ‘The Visible Archive‘ to investigate ways of visualising the holdings of the National Archives of Australia. Have you ever wondered what 360km worth of records looks like?

The collections of the NAA visualised by Mitchell's Series Browser.

This represents the holdings of the National Archives. Files within the archives are organised into series, and each square in this image represents a single series — there are about 60,000 of them. Naturally the size of the square gives an indication of the size of the series itself. It’s a fascinating and strangely beautiful picture.

It’s easy enough to pick out the World War One service records — Series B2455. In the interactive version of Mitchell’s series browser you can click on a box and display links between series, as well as other series created by the same government agency. Again, it’s not just a way of seeing the collection, but a means of exploring and interpreting it. As Mitchell says:

Visualisation enables us to literally show everything, to display large volumes of data in a way that reveals patterns and communicates context, but also provides access to the fine grain of individual elements.

But we can also employ such techniques to ask new kinds of questions. Can you imagine how Alexander Kelley and the other inhabitants of the Anzac Hostel must have felt in 1939? They had lost so much in the Great War, the ‘war to end all wars’, and yet within their own lifetime it was all happening again. More young men were answering the call, more lives were going to be destroyed.

There must have been a dreadful, disheartening moment when Australians realised that the Great War was not an end, but a beginning — the first in a series of devastating global conflicts. At some point the ‘Great War’ became the ‘First World War’, but when?

When did the 'Great War' become the 'First World War'?


This is one possible answer. This graph draws its data from the 50 million or so digitised newspaper articles in Trove, the National Library of Australia’s discovery service. It shows the proportion of newspaper articles that included the phrase ‘the great war’ compared to the proportion containing ‘the first world war’ (and variations thereof). The lines cross late in 1941. With German victories in Europe and Africa, the opening of the Eastern Front and the Japanese attack on Pearl Harbour, 1941 makes sense.

What is perhaps more intriguing is the dramatic peak in the occurrence of ‘the great war’ in 1939. It’s no surprise that the looming threat of a new conflict would provoke comment and comparisons, but it does make you wonder about the context of those discussions and how they might have changed as the reality of war edged closer.

To start exploring this I’ve harvested the content of the 6,600 articles from 1939 that included the phrase ‘the great war’. Using an online text analysis service called VoyeurTools I can quickly generate a picture of their contents.

This simple visualisation shows us the relative frequencies of words within the articles. It doesn’t reveal any great mysteries, but it does suggest some possibilities for further prodding. The prevalence of ‘time’ and ‘new’, for example — might these help us understand the shift in perspective from one war to the next? We can follow this up by browsing the different contexts in which the words were used.

But what actually is it that we’re actually searching? We know that Trove includes newspapers from 1803 to 1954, but if we’re really going to analyse shifting words and ideas it’s important to have a clear picture of the sources of those words.

Something like this perhaps. This graph shows the holdings of the Trove newspaper database on 4 August 2011, organised by state. You can see, for example, that if you’re searching on a topic between the 1920s and 1940s you’re probably likely to get more results from Queensland than anywhere else.

So starting from our location here, today, we can make connections across time and space. We can pull back and look at the big picture, or dive in and examine the fabric of a single life. Through the web we can build and explore a rich and complex contextual network.


It’s an exciting time to be a cultural data hacker. We now have a growing range of tools and technologies available for extracting interesting data from a wide variety of sources, both structured and unstructured.

The ‘Visible Archive’ project started with well-structured data, courtesy of Peter Scott, the developer of the Series System — the descriptive framework used by many Australian archives. But we’re rarely so lucky.

Even when the data starts off in nicely-organised fields in a database there’s no guarantee that that’s how it’s going to be delivered to our web browser. In order to extract the data from my Trove graphs, for example, I had to write a little program called a ‘screen scraper‘ to identify and save the important metadata elements from the raw web page itself.

Where there are no subject keywords we can infer them using techniques such as topic modelling. Where there are no access points we can identify people, organisations, places and events using special tools developed for named entity extraction. Where there are no common identifiers across datasets we can employ record linkage technologies to find possible connections.

We can count words, we can identify parts of speech, we can formulate a measure of the similarity of any two pieces of text. Once we have some useful data we can manipulate and enrich it. Place names can be geolocated — you simply send your place name off to a web service and get back its latitude and longitude.

Increasingly these sorts of tools are becoming accessible to anyone. For historians they offer a means of wrestling with rapidly-growing bulk of source material that is becoming available in digital form. How do you make use of 5 million digitised books, 50 million newspaper articles or the complete archive of every public message ever sent on Twitter?

The digital historian Dan Cohen has noted:

These computational methods which allow us to find patterns, determine relationships, categorize documents, and extract information from massive corpuses, will form the basis for new tools for research in the humanities and other disciplines in the coming decade.

Dan is involved in a number of interesting projects investigating the possibilities of these techniques — often grouped together under the heading ‘text mining’. One of these projects, ‘With Criminal Intent‘, is looking to see what patterns can be drawn out of the digitised proceedings of criminal trials held at the Old Bailey from 1645 to 1913. That’s 197,745 trials, in case you were wondering.

Here’s one of their visualisations showing how the length of trials varies over time. Much to the surprise of the research team, this graph suggests a dramatic shift in legal practice around 1825 — defendants started pleading guilty!

A visualisation by the With Criminal Intent project showing changing trial lengths.

Rather than falter under the growing weight of digital sources, these technologies can actually thrive. The more raw material available, the more chance there is to observe and track new patterns. As digitisation continues apace will we ever reach the point when history can simply be read from a graph?

There are some researchers at Harvard who seem to think that’s where we’re heading. Borrowing liberally from the store of scientific metaphors they have staked out the new field of ‘culturomics‘. By mining massive digital resources, like Google’s scanned books, they hope to map the ‘cultural genome’ that would enable us to follow the evolution of language and culture.

But there’s something quite barren in this ambition. I prefer the vision of digital humanist Stephen Ramsay, who commented in regard to the ‘With Criminal Intent’ project:

The Old Bailey, like the Naked City, has eight million stories. Accessing those stories involves understanding trial length, numbers of instances of poisoning, and rates of bigamy. But being stories, they find their more salient expression in the weightier motifs of the human condition: justice, revenge, dishonor, loss, trial. This is what the humanities are about. This is the only reason for an historian to fire up Mathematica or for a student trained in French literature to get into Java.

Ultimately it’s the stories that nourish, anger, inspire and depress us. The closely-packed map of places recorded in World War I service records is so powerful because we know that under each marker are men, women, families, communities — each with their own story. These new technologies offer new perspectives, they raise new questions, and they challenge us with new contexts to explore and understand. But there is still space for stories and perhaps we can use them to give our stories new life and depth.


This is another World War One service record. It belongs to Charlie Allen. Charlie enlisted three times in the AIF and was discharged on medical grounds each time. It seems he had a problem with his ankle.

Charlie’s service record notes a tattoo, proclaiming his love for ‘Maud Gordon’. He married Maud in Sydney in 1917 and had two daughters soon after.

Charlie survived the war without further injury, but was not so lucky in peace. On 11 March 1938, Charlie was crushed to death between two railway cars. The accident happened at the Bunnerong Power Station, only a short distance from his home in Matraville. He was buried nearby in the Botany Cemetery.

We also know quite a bit about Charlie’s early life. Why? Because Charlie’s father was Chinese and he was therefore categorised as a ‘half-caste’, as someone who was not white, and therefore fell under the restrictions imposed by the White Australia Policy.

Charlie was born in Sydney in 1896. His mother was Frances Allen (sometime sweet shop owner and brothel keeper), his father Charlie Gum (a buyer for Wing On company). Charlie was raised by his mother, but in 1909, at the age of 13, he was taken to China by his father.

NAA: ST84/1, 1909/22/41-50

This certificate granted Charlie an exemption to the Dictation Test. Without it, he may not have been allowed back into the country.

Every time one of many thousands of non-Europeans resident in Australia sought to travel overseas and return home again they needed one of these certificates.

Charlie’s father returned to Sydney, leaving him in China. He lived with relatives in the town of Shekki (inland from Hong Kong). Charlie was naturally homesick, but had no means of getting back to Australia. He wrote to his mother in 1910:

Do try and bring me home every minute I think of you and long for a piece of bread and butter this tucker is not doing me well.

His mother wrote to the Prime Minister Billy Hughes in an attempt to enlist government help but to no avail. Charlie finally returned to Australia in 1915.

Despite this experience, Charlie visited China again in 1922 for 7 months. Once again carrying papers to grant him re-entry to the country of his birth.

These fragments of Charlie’s life have been assembled by my partner, Kate Bagnall, a historian of Chinese-Australia. They are remarkable, and yet not so, because there are many thousands of stories like Charlie’s contained within the voluminous records generated by the administration of the White Australia Policy.

We’re all of course familiar with the general outlines of the White Australia Policy, and the way it underpinned conceptions of Australia as a nation in the first half of the 20th century.

But what we sometimes forget is that it was also a massive bureaucratic exercise.

Forms and certificates were printed, issued, used and filed. Regulations were modified, guidelines were distributed and administering officers were managed and advised. Individual cases were reviewed, policy was changed and new forms and certificates were printed, issued, used and filed…

Much of this system is now preserved in the National Archives.

You can get a idea of the range of material available from a case study Kate has prepared focusing on the efforts of Poon Gooey, a successful businessman in Horsham, to keep his wife and family in Australia.

If we look again at Charlie’s certificate from 1909 we can see that it contains a lot of interesting structured data:

  • name
  • place of birth
  • age
  • height
  • destination
  • date of departure
  • name of ship

We estimate that there are probably about 50,000 of these forms remaining in the Archives, and then there’s case files and a variety of other government documents.

Wouldn’t it be great if we could extract this structured data. If we could piece together the slivers of identity that remain within the Archives and give people back their lives.

This is the dream of Invisible Australians, a project Kate and I are trying to turn into a reality. Our aim is to build systems that will enable this data to be extracted, aggregated, shared and connected — whether to a family tree, a cemetery record, or another document in another archive.

Imagine being able to navigate the network of lives, families and relationships. To follow their journeys, to share their tragedies, to celebrate their small victories against a repressive system.

Imagine being able to watch them age.


We tend to assume that new technologies require us to change, to adapt. But sometimes they can take advantage of our strengths. Mitchell Whitelaw is interested in finding out what happens when you take large cultural datasets and try to ‘show everything’. Such an approach, he suggests, takes advantage of the raw processing power of computers, while giving us space to do what we’re good at — finding patterns, making connections, crafting meanings.

The History Wall tries to create a similar sort of space. The History Wall brings together material from a range of different sources — newspaper articles from Trove, biographies from the Australian Dictionary of Biography, records from a database of NSW convicts, population statistics, collection items from the National Museum of Australia — you can pretty much plug anything in as long as it has a date attached to it.

Irish History Wall

For a particular year, the Wall retrieves a random sample from the available sources, jumbles everything up and then throws it onto the screen. As a result, no two views of the Wall are ever quite the same. This is not a traditional exhibition. There is no curator controlling the content or designing the structure. It’s ephemeral, it’s serendipitous — instead of relying on an authorial voice to smooth over the gaps and transitions, it leaves open the cracks and allows new contexts to seep in and around each item.

As the pioneering digital historian Edward Ayers noted:

even isolated and inert pieces of evidence — a list, a letter, a map, a picture — can assume new and unimagined meanings when placed in juxtaposition with other fragments.

This is not an absence of narrative, but an opportunity for narration. Edward Ayers suggests that we’re actually quite comfortable filling in blanks and untwisting timelines:

Humans, presented with pieces of information about people, put things into the form of a story. They need not be simple stories, for we know how to deal with unexplained lapses of time, flashbacks, and overlapping narratives. We know how to imagine, infer, things happening at the same time in different places. Film and television train all of us at early ages to weave strands of narrative out of intentional (if carefully constructed) confusion and to take pleasure in that weaving.

And so I can show you a death notice, or a certificate and you will take those fragments, those isolated data points and you will construct a story — you will see the person behind them, you will imagine their life. It’s what we do. We’re good at it.

Computers on the other hand will just see data.

In her ode in praise of humanities data, digital humanist Amanda French wonders whether we always need to crunch our data into abstract, pliable forms:

What I wonder is whether instead we can begin with the data, or with a datum, and simply watch for what it may tell us, even if what it tells us is simply a story.

Yes we can. And we should teach computers how to do it as well. Not because we want them to take over. Not because they can necessarily do it faster or better. But because they can help us share, preserve and connect those stories.

Let’s think again about the array of documents that Kate has assembled to piece together the story of Charles Allen. How can you share this sort of material? Typically you’d ‘write it up’. You’d capture the story behind the data and commit it to words. The documents would then become evidence — points of connection between your text and the historical record.

So in order to share the meanings of these documents we remove them from the context of the person’s life and marshal them as allies to proclaim the authenticity of our rendering. Wouldn’t it be better if we could tell the story, but maintain within our texts the direct connections between sources and subject?

What we need is a data framework that sits beneath the text, identifying people, dates and places, and defining relationships between them and our documentary sources. A framework that computers could understand and interpret, so that if they saw something they knew was a placename they could head off and look for other people associated with that place. Instead of just presenting our research we’d be creating a whole series of points of connection, discovery and aggregation.

Sounds a bit far-fetched? Well it’s not. We have it already — it’s called the Semantic Web.

The Semantic Web exposes the structures that are implicit in our web pages and our texts in ways that computers can understand. The Linked Data movement takes the basic ideas of the Semantic Web and turns them into a collaborative activity. You share vocabularies, so that other people (and computers) know when you’re talking about the same sorts of things. You share identifiers, so that other people (and computers) know that you’re talking about a specific person, place, object or whatever.

Linked Data is Storytelling 101 for computers. It doesn’t have the full richness, complexity and nuance that we invest in our narratives, but it does at least help computers to fit all the bits together in meaningful ways. And if we talk nice to them, then they can apply their newly-acquired interpretative skills to the things that they’re already good at — like searching, aggregating, or generating the sorts of big pictures that enable us to explore the contexts of our stories.

This is why we’ve always imagined Invisible Australians to be something more than an online database. We want to provide points of connection that other people can build into their own stories. But to do that we have to pay attention to things like vocabulary management and authority control, we have to construct web addresses that are not going to break every time we upgrade our software. We have to think about the sorts of things we’re talking about — not just people, but government agencies, legislation, certificates, and correspondence. How do we describe these entities and what sorts of relationships do they have?

And of course we need to expose all these structures so that we can say, these things are people, these are events, these are places and these are documents.

Or perhaps, to introduce Alexander Kelley.

Or remember Charles Allen.


You might be wondering why we don’t just leave it all to the computers themselves. Didn’t I just talk about all the exciting new tools and techniques that enable us to analyse the structures of texts? Perhaps we should just wait for the Culturomics guys to solve all the problems.

But who defines the problems?

Our postmodern sensibilities encourage a suspicion of neutrality. Labels like ‘the new museology’ or Archives 2.0 reflect an awareness that the way we describe and arrange our collections is itself culturally-determined. It’s not just a matter of what our descriptive systems show, but what they hide.

Tim Hitchcock, another member of the ‘With Criminal Intent’ team, has described how online technologies can change the way we access archives. Instead of being forced to navigate the hierarchical structures that archives impose on records, which in turn tend to reflect the workings of the institutions that created the records, we can directly find the people whose lives were regulated, influenced, shaped or controlled by the policies of those institutions.

Instead of merely hearing ‘the institutional voice… in all its stentorian splendour’, he says, we can listen in to ‘the quieter tones uttered by the individual’.

This reminds us that search boxes, along with other digital tools, themselves embody arguments. There are assumptions built into their code about what is relevant, what is significant, what is necessary.

We can build our own tools of course, and we can critique other people’s algorithms. But what if we just want to collect and share stories?

Linked Data gives us a way to present an alternative to Google’s version of the world. We can argue back against the search engines, defining our own criteria for relevance, and building our own discovery networks.

Changing the way we access resources changes the sorts of stories we can tell. Tim Hitchcock asks:

What happens when institutions and archives are ‘decentred’ in favour of the individual? What changes when we examine the world through the collected fragments of knowledge that we can recover about a single person, reorganised as a biographical narrative, rather than as part of an archival system?

Perhaps the invisible become visible.

the real face of white australia

In many of the presentations I’ve given in recent times I’ve managed to include a question raised by Tim Hitchcock in his chapter in The Virtual Representation of the Past. Tim asks:

What changes when we examine the world through the collected fragments of knowledge that we can recover about a single person, reorganised as a biographical narrative, rather than as part of an archival system?

The idea of turning archival systems on their head to expose the people rather than the bureaucracy is what motivates Kate Bagnall and I in our attempts to make the Invisible Australians project into a reality.

Invisible Australians aims to liberate the lives of those who suffered under the restrictions of the White Australia Policy from the rich archival holdings of the National Archives of Australia and elsewhere.

We always knew that the portrait photographs, included on a range of government documents, would provide a compelling perspective on these lives, but we weren’t quite sure how we were going to extract them. Up until last weekend, I’d assumed that we’d develop a crowdsourcing tool that contributors would use to mark-up the photos.

Now I’m not so sure.

In the space of a couple of days I’ve extracted over 7,000 photographs and built an application to browse them — here is the real face of White Australia

How did I do it? Paul Hagon, at the National Library of Australia, gave a presentation last year in which he explored the possibilities of facial detection in developing access to photographic collections. The idea lodged in my brain somewhere and a few days ago I started to poke around looking to see how practical it might be for Invisible Australians.

It didn’t take long to find a python script that used the OpenCV library to detect faces in photographs. I tried the script on a few of the NAA documents and was impressed — there were a few false positives, but the faces were being found!

So then the excitement kicked in. I modified the script so that instead of just finding the coordinates of faces it would enlarge the selected area by 50px on each side and then crop the image. This did a great job of extracting the portraits. I tweaked a few of the settings as well to try and reduce the number of false positives. Eventually, I developed a two-pass system that repeated the detection process after the image had been cropped and it’s contrast adjusted. This seemed to weed out a few more errors. You can find the code on GitHub.

Once the script was working I had to assemble the documents. I already had a basic harvester that would retrieve both the file metadata and digitised images for any series in the NAA database. Acting on Kate’s advice, I pointed it at series ST84/1 and downloaded 12,502 page images.

All I then had to do was loop the facial detection script over the images. Simple! The only problem was that my 3-year-old laptop wasn’t quite up to the task. As it’s CPU temperature rose and rose, I was forced to employ a special high-tech cooling system.

Keeping my laptop alive...

But after running for several hours, my faithful old laptop finally worked it’s way through all the documents. The result was a directory full of 11,170 cropped images.

The results

There were still quite a lot of false positives and so I simply worked my way through the files, manually deleting the errors. I ended up with 7,247 photos of people. That’s a strike rate of nearly 65% which seems pretty good. The classifier, which does the actual facial detection, was probably trained on conventional photographs rather than on the mixed-format documents I was feeding it.

Then it was just a matter of building a web app to display the portraits. I used Django for the backend work of managing the metadata and delivering the content, while the interface was built using a combination or Isotope, Infinite Scroll and FancyBox.

It’s important to note that the portraits provide a way of exploring the records themselves. If you click on a face you see a copy of the document from which the photo was extracted. A link is provided to examine the full context of the image in RecordSearch. This is not just an exhibition, it’s a finding aid.

What next? There are many more of these documents to be harvested and processed (and many more still yet to be digitised). I will be adding more series as I can (though I might have to wait until I can afford a new computer!). I’d also like to explore the possibilities of facial or object detection a bit more. Could I train my own classifier? Could I detect handprints, or even classify the type of form?

In the meantime, I think our experimental browser helps us to understand why the Invisible Australians project is so important — you look at their faces and you simply want to know more. Who are they? What were their lives like?

UPDATE: For more on the photos and the issues they raise, see Kate Bagnall’s posts over at the Tiger’s Mouth.

Liberating lives: invisible Australians and biographical networks

Presented at the Life of Information Symposium, 24 September 2010.
Slides are available on Slideshare.

Charlie Allen's palm print
This palm print belongs to a 12-year-old boy called Charlie Allen.

Charlie was born in Sydney in 1896.

His mother was Frances Allen (sometime sweet shop owner and brothel keeper), his father Charlie Gum (a buyer for Wing On company).

Charlie was raised by his mother, but in 1909, at the age of 13, he was taken to China by his father.

His father returned to Sydney, leaving Charlie in China. He lived with relatives in the town of Shekki (inland from Hong Kong) for 6 years.

Charlie was homesick, but had no means of getting back to Australia. His mother attempted to enlist government help but to no avail. Charlie finally returned in 1915.

The following year he enlisted in First AIF (well actually he enlisted three times, and was discharged as medically unfit each time).

Charlie married in Sydney in 1917 and had two daughters soon after. He returned to China in 1922 for 7 months.

Charlie Allen died in 1938 as the result of an industrial accident. He was 41.

How do we know all this about Charlie Allen?

We know this because there are fragments of Charlie’s life scattered throughout the holdings of the National Archives of Australia.

The CEDT from 1909 when he left Australia with his father:

Charles Allen 1909 - CEDT front

NAA: ST84/1, 1909/22/41-50


A letter from his mother to Prime Minister Billy Hughes, seeking help to return Charlie to Australia:
Letter to Billy Highes from Charlie's mother.

NAA: A1, 1911/13854


His WWI service record:
Charles Allen's WWI attestation form

NAA: B2455, ALLEN C A


An identity form relating to his trip to China in 1922:

NAA: SP42/1, C1922/4449


But of course Charlie is not alone in the archives.

Charlie’s father was Chinese, he was therefore categorised as a ‘half-caste’, as someone who was not white, and fell under the restrictions imposed by the White Australia Policy.

The certificate from 1909 granted Charlie an exemption to the Dictation Test. Without it, he may not have been allowed back into the country.

Every time one of many thousands of non-Europeans resident in Australia sought to travel overseas and return home again they needed one of these certificates.

We’re all of course familiar with the general outlines of the White Australia Policy, and the way it underpinned conceptions of Australia as a nation in the first half of the 20th century.

But what we sometimes forget is that it was also a massive bureaucratic exercise.

Forms and certificates were printed, issued, used and filed. Regulations were modified, guidelines were distributed and administering officers were managed and advised. Individual cases were reviewed, policy was changed and new forms and certificates were printed, issued, used and filed…

For example, between 1901 and 1911, 400 circulars were issued to port officers about immigration restriction. The confidential manual on immigration restriction grew from one page in 1902 to more than 200 in 1912.

Much of this system is now preserved in the National Archives.

For the years between 1902 and 1948 there remain:

  • More than 50,000 CEDTs
  • 90 shelf metres of records
  • 15,000 case files

And within those many thousands of files are the scattered fragments of lives such as Charlie’s — lives that were controlled, monitored and documented in a vain attempt to make Australia ‘white’.

We’ve already seen today some wonderful examples of how these fragments, these slivers of existence, can be found, extracted, aggregated and displayed. But I think it’s worth considering for a moment what happens when we do this.

The historian Tim Hitchcock, behind projects such as the Old Bailey Online and London Lives, has reflected on the impact of digitisation on our access to archives. Archives, he notes, tend to reflect the assumptions and practices of the institutions that created them.

But by providing new ways into these records systems, technology can undermine the power relations that persist within their structures.

‘What changes’, asks Tim Hitchcock, ‘when we examine the world through the collected fragments of knowledge that we can recover about a single person, reorganised as a biographical narrative, rather than as part of an archival system?’

I don’t know, but I think we should find out, don’t you?

**********

I hope you’ve all collected a mini card. These themselves provide a little glimpse at the real face of White Australia and I’d invite you all to head over to the National Archives website, do battle with the monster that is RecordSearch, and look up the file references that are on each card.

The cards are part of a project that Kate Bagnall and I are trying to develop — Invisible Australians.

I should note too that the cards, and most of the examples I’m showing you here today are the product of Kate’s long and detailed research into Chinese-Australian families. In modern project management parlance, Kate is the domain expert, while I am merely the technical resource.

If we look again at one of the CEDTs, we can see that there’s a lot of useful structured data:

  • name
  • place of birth
  • age
  • height
  • destination
  • date of departure
  • name of ship

Invisible Australians has the modest aim of extracting this data from the 50,000+ forms in the National Archives. But of course that’s just the start, because each person might have used a number of certificates — so then it’s a matter of matching these identities.

Invisible Australians

http://invisibleaustralians.org

And then there are a range of other related forms, not to mention case files, alien registration documents, naturalisation applications…

Obviously we can’t do it alone. We’ll be creating a crowdsourcing tool to extract and link the data.

It’s ridiculously ambitious, totally unfunded and is likely to take over our lives.

Is it worth it?

Imagine being able to navigate the network of lives, families and relationships. To follow their journeys, to share their tragedies, to celebrate their small victories against a repressive system.

Imagine being able to watch them age.

Pauline Ah Hee and Shadee Khan

Is it worth it? We think so.

**********

For Tim Hitchcock technology opens up the possibility of writing a new history from below, exploring how the poor, the marginalised and the powerless navigated the institutions of the modern state. But it’s not just about search engines and databases. He talks about making ‘best use of the technology of emotions and representation — how you use words and pictures and a story to impact, not just on what people think, but what they see in their mind’s eye’.

In this project, the photos matter. I hope the irony in our project title is obvious.

Some of the faces of Invisible Australia

This is the real face of White Australia.

The photos remind us that the project is not just about shifting data around — these are lives, these are people.

But this brings its own challenge, for if we are seeking to liberate these lives from the fragmentation and obscurity of bureaucratic systems then we should be asking what are we liberating them into?

A database?

This is not just an exercise in data creation and management. We also have to think carefully and creatively about issues of representation, access and discovery.

We have to give these lives back their freedom to associate, to have relationships, to make connections.

We need to embed these lives in a variety of contexts and combinations. To make room for serendipity, celebration, sadness, and yes, even play.

We need to bring these lives into a rich and ongoing conversation with the world.

But how?

**********

I’ve been working on a little experiment for the National Museum of Australia called The History Wall. What the History Wall does is quite simple, it pulls together data on the fly from a variety of sources including People Australia, the Australian Dictionary of Biography, the National Library’s newspapers project, historical population data from the Bureau of Statistics, photos from the Flickr accounts of the PowerHouse Museum and the National Archives, and the collection database of the National Museum itself. It chooses randomly from all this stuff, throws the results up into the air and then displays them however they happen to fall. No two views are ever quite the same.

The History Wall

http://defining.net.au/wall/

It’s something more than a timeline. To me it’s more like a celebration of context and serendipity. There’s a richness to it, a sense of discovery and fun, but there’s also fragility — next time you look it might be gone.

It’s a bit like history itself.

It’s a bit like the world.

How do we create spaces for our data to merge and mingle? How do we encourage the development of new contexts and connections?

I think the first thing we have to do is stop thinking about databases and dictionaries, registers and encyclopaedias. Don’t get me wrong, I’m not being critical of the wonderful projects we’ve seen today. I just think we can use all this work better if we stop thinking about individual resources and start developing on a web scale, on a global scale.

Yes, we have the technology. Time today has spared you from a detailed discourse on the Semantic Web, but I do want to focus on one aspect.

You may have heard of Linked Data, it’s a set of guidelines to help you publish your data to the Semantic Web. There are only four basic principles and I’m only going to talk about one of them. It’s one of those deceptively simple things. You look at it and think, ‘yeah, ok’, but before too long it’s starting to turn your brain inside out.

Use URLs to identify things in the real world.

Yeah, ok…

You know what URLs are, web addresses, the things you type in your browser’s location field.

And hopefully you know what things in the real world are: people, places, objects, events, ideas…

Now you may have detected a problem here, because no matter how many times you click the refresh button, your web browser is not going to be able to use such a URL to magically deliver you the real world thing.

Well, unless you’re on eBay.

Fortunately, the Linked Data guidelines provide for a bit of technical trickery that allow your browser to retrieve not the real world thing, but some information about that thing — perhaps in the form of a web page.

Why bother?

Names are powerful.

We share and use names to talk about things. Computers are the same. If we use URLs to identify things in the real world, then computers can start talking about them.

We can define and explore real-world relationships in an online environment. We can create rich, meaningful linkages across databases, across disciplines, across the world.

We can start building and thinking on a web scale.

**********

Thanks to the People Australia project, I can confidently claim that this is me:

http://nla.gov.au/nla.party-479364#foaf:Person

I keep meaning to get it on a t-shirt.

The most exciting thing about People Australia is not the EAC records or the aggregation of resources — it’s the identifiers, because they enable us to say things about people anywhere on the web that computers can understand and relate back to a specific real world entity — a person.

You can start doing it now with Wragge’s Identity Browser.

Wragge's Identity Browser

http://wraggelabs.com/identities/

This is a little tool I built using the People Australia API. It makes it easy to find identifiers for people and organisations, and it supplies you with some code that you can drop into a blog post or web page that will tell a computer that a name relates to a thing called a ‘person’ , that this person’s name has a certain standard form, and that this person can be uniquely identified by People Australia.

Even if you don’t publish a website or a blog, you can use People Australia identifiers to build semantic linkages. Wragge’s Identity Browser also creates machine tags for you. Machine tags are like normal tags but with built in semantics. When coupled with identifiers they enable you to do some pretty powerful things.

You could for example use machine tags in Flickr to tell computers that a certain photo depicts a person uniquely identified by People Australia. In fact, people have been doing just that.

Flickr Machine Tag Challenge

http://wraggelabs.com/fmtc/

The Flickr Machine Tag Challenge is a sort of scoreboard that I built to encourage people to start adding People Australia enriched machine tags to photos. More than 1200 tags have been added to over 1000 photos. Feel free to join in!

The point is that the technologies already exist to enable us to build web scale biographical resources. Not dictionaries or databases as we know them, but networks capable of constant expansion, elaboration, and cooperation.

What we need are more tools to make it simple, recipes to make it obvious, examples and applications to make it popular, and leadership to make it all seem possible.

**********

Of course most of the lives we hope to liberate through Invisible Australians will not be represented in People Australia.

Not yet.

But Invisible Australians will offer a point of aggregation and disambiguation that will enable our people to find their way from the bureaucratic recesses of the White Australia Policy to a place on the national stage.

And we will encourage others to do likewise. Basil can’t do all the work. The centralised system has to be fed through centres of aggregation and collaboration.

Similarly, there are many great resources already out there relating to Chinese-Australians. There are hordes of family and local historians compiling and publishing biographical data. We want to identify people in these resources and link to them.

We want to publish up to People Australia and link down to a single headstone in a lonely country cemetery.

But to do this we need to help people make their resources linkable. To help them create persistent, re-usable URLs, and expose their data in standard formats. To create Linked Data, even if they have no particular interest in the Semantic Web.

Invisible Australians

http://invisibleaustralians.org/

Invisible Australians is not just about extracting data from archives. It’s also about working with others to build capacities and demonstrate possibilities.

It’s ridiculously ambitious, totally unfunded and is likely to take over our lives.

Is it worth it?

We think so.