The end users of AHAD are intended to be:
- Historical Archaeologist researchers
- Private-sector consultants
- Curators/Collection Managers
- Government Archaeologists/Administrators
Some of the things we intend for them to be able to do are:
- Upload their own datasets
- Share datasets with their contemporaries
- Create their own ontologies & coding sheets
- Integrate their datasets with those of their contemporaries
- Search in datasets for specific terms
The application we have selected (tDAR) allows for all of these functions and more. We will, of course, have to add some functionality to the system… but the application’s open source model means that we can contribute additional functions back to the original developers.
The primary features of the application we intend to use are:
- Ability to import existing datasets (artefacts, stratigraphy, images etc.)
- Ability to search multiple datasets
- Seamless integration of artefact, stratigraphic and occupancy data (object–place–people)
- Customisable datasets
- Flexible lexicon (synonyms, ontologies, coding sheets)
- Multi-user, editable environment
- Multi-dimensional: accommodate basic, advanced and highly specialised datasets
- Sync with other online datasets
The features we will need to add are:
- Adding OAI-PMH to tDAR and installing another OAI-PMH as a transforming proxy to convert tDAR’s own schema into RIF-CS.
- Introducing open LDAP as an authentication provider and writing a tDAR component to the LDAP server.
- Adding a module to tDAR to use the ANDS PIDS service to create identifiers.
- Adding mandatory copyright properties and other licensing and attribution data
- Making additions to allow for rebranding, including alterations to system menus and export of headers, footers and licensing information
- modification to metadata editing screens to integrate new functionality
Investigating currently available online archaeological databases has yielded some interesting finds, and I have some observations about the various different methods of data acquisition used by existing systems.
Opencontext.org currently provides 25 integrated tables with a granular, faceted filtering system and over 200,000 artefact image records. It has a fairly accessible interface and can also send data over ATOM/RSS via a RESTful interface.
The data offered by opencontext.org appears to be pre-normalized which is something that may present us with a problem: pre-existing datasets need to be altered to conform with the OpenContext schema. While this is certainly possible sometimes (and in some cases automation is an option), at other times the data cannot be made to match easily, fields in the original dataset are more or less granular than those required by the OpenContext schema, meaning is ambiguous or the researcher used a coding system without hard limitations, making it hard to match and verify their data with a controlled vocabulary. Also, the copyright owners would likely have to give written permission to allow integration to begin with, and that presents some potential legal issues.
tDAR.org currently contains over 300,000 documents and over 300 datasets. It is superficially similar in concept to Opencontext, but places a lot more emphasis on dataset flexibility and seems to require a much greater involvement by the researcher or uploader at the point of data ingest. The reason for this requirement is simple but important to understanding the intended purpose of the application: tDAR allows researchers to upload datasets without the need for pre-normalization, map their columns to common ontologies and integrate their own datasets with those of other researchers. This is, of course, in addition to storing the datasets and related metadata for review by interested parties.
While it’s not terribly relevant when reviewing their functionality, it’s worth noting that neither of these services presently have very much Australian archaeological data available.
Welcome to the project development blog of the Archaeology Database project!
The project is an effort of the Archaeology Program of La Trobe University, Melbourne, and is funded by the Australian National Data Service, through their “Seeding the Commons” program. The aim of the project is to build online data infrastructure for archaeology in Australia, to provide a public home for archaeological datasets, and to promote their reuse.
My name’s Conal Tuohy. I work for the Victorian eResearch Strategic Initiative, and I’m the technical lead on the project.
In 2010 the Faculty of Humanities and Social Sciences and the eResearch Office, in partnership with Victorian eResearch Strategic Initiative (VeRSI), received funding from the Australian National Data Service (ANDS) to build a national online database, the Australian Historical Archaeology Database (AHAD), for historical archaeological catalogue data and associated stratigraphic and historical records.
Each year archaeologists (both academic and private consultants) excavate tens of thousands of artefacts from historical archaeological sites across Australia. While some states (e.g. Victoria) require catalogues to be prepared in a standard format, the majority of catalogue data are stored in small, standalone spreadsheets or custom-built databases, and few are made freely available. There is no central register of these individual datasets, and many significant collections are simply unknown to archaeological researchers.
Between 2001 and 2004, the La Trobe led ARC-Linkage project Exploring the Archaeology of the Modern City (EAMC) created two research databases which offered, for the first time, a central database of 700,000 artefacts from multiple historical archaeological sites and a companion dataset of historical occupancy data relating to 2,200 individuals who occupied those sites. While extensive in their data content, the databases themselves are limited in their structure and require significant design input to make them truly effective tools for managing and sharing historical archaeological data.
Funding has been provided by the Seeding the Commons program of the Australian National Data Service (ANDS) to redevelop these two EAMC databases into an online platform and storage house for searching, entering and managing historical archaeological data; and seek new datasets from the private, public and tertiary sectors. The final product, the Australian Historical Archaeology Database (AHAD), is due for release in December 2011 will enable researchers to access a vast dataset that is currently unavailable to them, and provide the platform for future datasets to be made freely available in a standardised fashion. In addition, it will enable automated registration of datasets in Research Data Australia.