Projects in focus
The Dictionary of Sydney
ARC Linkage Project
Original historical writing - entries - commissioned by the Dictionary editors are enriched with annotations linking to a range of historical resources (images, people, places, events, references). The Dictionary is unique in being built around an explicit ontology of historical entities and a rich network of annotated relationships between entities. This structure defines a dynamic, navigable network of history rather than locking content into a set of static web resources or a relational database structure.
The Dictionary breaks new ground in historical content creation and editorial workflow, interactive website publishing and the automatic building of web pages from multiple resources (including entries, images, map and timeline data, records of people, places, structures events and so forth). The Archaeological Computing Laboratory’s unique academic content management system (CMS) – Heurist (heuristscholar.org) – provides a collaborative web-based editorial platform in which all of the information can be collected, connected and annotated. The web site is generated through dynamic queries for testing, then published so that it can be served by a commercial data centre independent of the CMS. This generates a fast and stable public interface. The web site is generated through dynamic queries for testing, then independent of the CMS. This generates a fast and stable public interface.
Ontologies and relationships. The project takes a much deeper look and information architectures than would be necessary for simple web publishing. The aspiration is to define and publish a model for the recording of history that is richly descriptive but elegantly simple thus making it easy to implement, query and archive. All data structuring work done in Heurist to support project workflows has been mirrored in industry standard semantic modelling methodologies and so at the end of the first stage of the project we not only have the public website but a fully expressed OWL ontology and the ability to export all or part of the record set as RDF.
Heurist and Cocoon. The Dictionary of Sydney is built around the Heurist collaborative academic database. An editorial interface (right) allows the content editors to annotate and link the entities (records) which constitute the Dictionary. The web site is generated through XML feeds which are formatted using XSL stylesheets through Apache Cocoon pipelines.
The Sydney Journal, a peer-reviewed journal for historical writing published by the Dictionary of Sydney Trust.
The Dictionary of Sydney Trust. The Dictionary of Sydney has been constituted as a charitable trust with a Board of Trustees whose initial membership included the CIs of the first Linkage grant and representatives of the City of Sydney, Historic Houses Trust, State Library of NSW and NSW Archives. The Board has secured funding to continue work on content commissioning and editorial control.
Gallipoli: The First Day
ABC Innovation Web Application
The ABC wanted to use a 3D environment to enable web users to gain a deeper understanding of what it was like to be at the Gallipoli landings in 1915. They also wanted a way to show the great variety of archival content (photography, diaries, oral history etc) in context. The DIU were able to provide the ABC significant assistance in the structuring of the many thousands of separate records that make up the collection. We also worked closely with Harvey Broadbent from Macquarie University to ensure that maps and other material were correctly sourced and appropriate. Using Heurist as the backend for the project gave the ABC a new way to look and information. We were told that they couldn’t have done the project without Heurist and in the end the site was awarded the inaugural AFI Screen Innovation Award.
The database that sits behind the public site is a highly related graph like structure with no built in hierarchy. There is no specific starting point and no fixed navigational path. The way this is integrated with the high quality 3D graphics and the intelligent interface design means users can discover their own pathways thus adding immeasurably to the learning experience.
While the ambitious final output was unusual this project began in a similar way to many others with standard systems analysis procedures to determine the correct data model. ABC staff also had to be introduced to the Heurist way of doing things – specifically, flexible record types and an “everything is related” approach to the world. Fairly soon into the project the team at the ABC caught on and started entering vast quantities of event, person and multimedia records and linking them all together. We were not involved in producing any final output but instead the corporation relied on the sophisticated XML output methods provided by Heurist and in the process we got to test and refine these.
A major public history website and an ongoing research resource in the form of Heurist. The ABC were awarded the AFI’s inaugural Award for Screen Content Innovation in 2009. The award citation states:
The AFI Award for Screen Content Innovation is an award in recognition of creativity within the screen industry. This award showcases the successful employment of new media tools to either extend traditional formats or develop new formats, and in so doing create more rewarding and engaging audience experiences.
Delivering storytelling in a pioneering three dimensional documentary style, Gallipoli: The First Day empowers users to navigate sophisticated narratives, creating their own unique experiences.
Digital Harlem 1915-1930
ARC Discovery Project
The project is producing an ethnographic portrait of Harlem in 1915-1930 when the arrival of migrants from the south and the West Indies transformed the neighbourhood. The basic dataset is surviving case files of the Manhattan District Attorney, which reveal all manner of things that would not ordinarily be labelled criminal - street life, black language, music, family life - as well as evidence of the role of gambling, violence and confidence men in the black community. In addition, it uses probation files, prison records, undercover investigations, social surveys, census schedules and the two major newspapers published in Harlem.
The site employs digital technology to integrate a range of sources to visualize and explore the spatial dimensions of everyday life in Harlem. The on-line database that supports the site enables data to be entered from any web-accessible location. Queries against this data are then used to generate interactive map displays dynamically. The database can be searched for, and the interface can then map, particular places or types of location, events, where an individual lived his/her life, or moments in time – or any combination of such data. Both the database and the map interface were refined over a period of more than three years as data was added, additional types of data were integrated and the researchers developed a clearer sense of how the data could be analysed and visualised.
The MySQL database was developed using the Archaeological Computing Laboratory’s T1000 database creation tool. The address locations recorded in the historical sources were initially mapped using an on-line geo-coding service and displayed in Google Maps. These locations were refined and additional locations added with reference to Bromley Fire Risk maps of the study area from the 1930s. In 2010 the maps from a volume of the Bromley maps was digitised and geo-registered and delivered in Google Earth to provide a contemporaneous background for the historical data.
In 2010, the American Historical Association awarded Digital Harlem the inaugural Roy Rosenzweig Fellowship for Innovation in Digital History. The citation notes that the site:
"presents the social history of a particular time and place in an elegant way that encourages exploration and new discoveries. The team behind the site draws on the strength of the primary sources and uses digital techniques to allow the viewer to see elements and patterns during the Harlem Renaissance that would be difficult to characterize in narrative."
"In addition to being open access, Digital Harlem is open-ended: researchers can explore themes of interest to them, layering experimental searches upon each other to envision the character and interactions of everyday life. The site also powerfully shows what can be done with the combination of common technology (Google Maps) with deep archival research and outstanding web design and functionality."
The History of Balinese Painting
ARC Linkage Project
Due in part to the work of Margaret Mead and Gregory Bateson in the 1930's a body of scholarship exists around the significant collections of traditional and more recent Balinese paintings held in the Australian Museum, the American Museum of Natural History, Princeton University, several important Dutch institutions and in various other significant collections around the world. This project is about using Heurist to significant aid scholars of the genre by building a virtual collection referencing all these works and cross referencing them to profiles of artists who painted them, locations where they were painted and the history of their ownership.
The project makes extensive use of the Heurist research infrastructure not only in its basic data model but also for its use of XML/XSL transformations to produce a high quality browsing display where a researcher can navigate through related content and thus gain a deeper understanding of the context of a particular work. Heurist benefits from a flexible data structure that allows new record structures to be easily defined. The can then be shared with other Heurist installations along with associated data entry forms and display templates. This facility for reuse saves effort and makes record structures combatively across databases. The project has taken many data structures from the shared pool while the deeper understanding of artwork related metadata gained has enabled the development of new artist and artwork record types.
Steven Hayes has worked in the role of systems analyst for the project. Essentially we came to the project with a mostly complete tool set which has been developed to support or many other projects. The system analyst job was largely to conceive of how to model this particular knowledge domain with already existing components if available and to define new information objects as required. As always there has been an element of training and presentation required to enable users and project stake holders to “get their head around” the Heurist way of doing things.
A well thought out, comprehensive, accessible and sustainable research resource. There is an intention to publish a book with some of the content of the database and that should be easily accommodated via Heurist and the Cocoon XML publish framework that sits alongside it.