Research

Seeing Triple: Archaeology, Field Drawing and the Semantic Web

My PhD research centred on how archaeological field data, and in particular the spatial information held in field drawings can be made part of the Semantic Web. The World Wide Web has profoundly changed the way we acquire, use and share information. At the same time, the growth of the Web has been hampered by its structure, which is made up of a 'Web of Documents'. This Web of Documents has been described as being 'a mile wide, but an inch deep' and using the Web can be a chaotic, unfocussed (albeit sometimes serendipitous...or downright funny) trawl through vast amounts of information to get to anything useful. The development of the Semantic Web is an attempt to create a deeper, more meaningful Web, by changing the Web of Documents into a 'Web of Data'. Freeing data from the Web documents and traditional, proprietary databases allows it to be combined, manipulated and presented in new ways, thereby making it possible to ask new and different questions, across multiple heterogenous data sources. Spatial data presents particular challenges for Semantic Web developers, but these challenges will have to be addressed for archaeologists and others whose research has a large spatial component. 

This thesis explores the Semantic Web with relation to archaeology, and whether it is yet possible for non-specialist archaeologists to create, use and share their data using Semantic Web technologies and principles. It also considers whether spatial data derived from field drawings can be incorporated alongside textual data, to ensure a more complete archaeological record is represented on the Semantic Web. To determine if these two related questions can be answered, a practical application was undertaken, followed by a discussion of the results, and recommendations for future work.

Two archaeological datasets were chosen for the practical application. The first was an Anglian and Anglo-Scandinavian site in the Yorkshire Wolds located near Burrow House Farm, Cottam, excavated by the Department of Archaeology at the University of York. The second was from the Anglo-Scandinavian area of the multi-period Hungate site in the York city centre, excavated by the York Archaeological Trust. One of the primary tenets of the Semantic Web is interoperability of data, and the sites were chosen because they were related archaeologically, but differed technologically. Both datasets included field drawings from which data could be extracted, along with augmentory databases to enhance the demonstration. The data was carried through a complete workflow, from extraction, alignment to an ontology, translation into RDF, querying and visualisation within an RDF store, and through to publication as Linked Data.

This practical application was completed primarily using newly available generic tools, which required a minimal amount of specialist knowledge during most phases of the process. It demonstrated it is currently possible for non-specialist archaeologists to work with their data using Semantic Web technologies, including some data derived from field drawings. It showed how the Semantic Web allows archaeologists to use their data in new ways, and that it is a fruitful area for further work.

This thesis is available via White Rose eTheses: http://etheses.whiterose.ac.uk/2194/

Scalable Vector Graphics: An Exploration of Potential Archaeological Applications

My MSc dissertation, Scalable Vector Graphics: An Exploration of Potential Archaeological Applications, was an analyses of SVG and its potential as a tool for bringing vector based archaeological information to the Web. It is an overview of the various ways archaeologists use vector graphics for the analysis, interpretation and communication of spatial data, and a discussion of the particular challenges associated with presenting that information on the Web. It outlines the characteristics and properties of SVG, including what it is, what it does and why it might be a useful tool for archaeologists. It also includes a discussion of future trends for SVG and its potential as a tool for archaeology. It explores not only how it is used for electronic publication on the Web, but also how it may be incorporated into excavation and post-excavation.

To demonstrate how SVG works, I created a practical implementation to illustrate how it might be used to publish vector-based archaeological information on the Web. Three plans and three section drawings were taken from Jeremy Haslam's Excavations at Cricklade, Wiltshire, 1975, which were published in Internet Archaeology (Issue 14, 2003), and digitized for online publication by Guy Hopkinson. The goal of the exercise was to take this set of images, created in AutoDesk's AutoCAD and Adobe Illustrator and interpret them online using SVG, whilst maintaining the same functionality built into them by Hopkinson. The case study includes an overview of the Cricklade project and information about the vector based plans and sections to be converted into SVG format. It also includes a discussion of the purpose, execution and findings of the demonstration.

A distillation of this research was published in 2006 as Archaeological Vector Graphics and SVG: A case study from Cricklade in Internet Archaeology (Issue 20).

Profile

• University of York Archaeological Information Systems research group
• Reviews Editor for Internet Archaeology
• EU Marie-Curie CHIRON Research Fellow
• Research Manager for the Department of Archaeology