9. Archaeological metadata example

Take a look at this picture:

Very pretty, isn't it? Well, wouldn't you like to know what it is, where it is from, and what it relates to? Indeed. Precisely the reason for metadata. So here is the Dublin Core metadata for the digital data behind this image:

Element Name Type Element Description
Title - Fluxgate Gradiometer Survey of Whitrighill
Author or Creator Fieldwork Director Kate Clarke
Image Processor Alicia Wise
Subject and Keywords-Whitrighill
-Iron Age
- Geophysics
Description - Results from 1991 survey of the Whitrighill site in the Borders, Scotland. Data collected using a Geoscan FM-18 gradiometer.
Publisher - Internet Archaeology
Other Contributors Project Newstead Research Project.
Funding Agencies National Museum of Scotland
University of Bradford
Borders Regional Council
British Academy
Society of Antiquaries of London
Date - January 1996
Resource Type - Geophysical survey image
Format - Gif
Resource Identifier - whitmag1.img
Relation Excavation Reports Borders Regional Council
Aerial Photograph National Monuments Record of Scotland BW5154
Geophysical Survey Newstead Research Project
Aerial Photograph Interpretation Newstead Research Project
Source - Gif file created from bitmap image generated from Contors version 1995 from geophysical survey data collected in 1991
Language - English
Coverage - 622000 345000
- Borders Region
- Scotland
- Roman Iron Age
- 300 BC - 200 AD
Rights Management - Copyright by the Newstead Research Project. Please do not copy or re-use this image without the prior permission of Dr. Rick Jones, Department of Archaeological Sciences, University of Bradford, Bradford BD7 1DP, UK.

Table 2: Dublin Core Metadata for Whitrighill Geophysical Image

This metadata record provides you with enough information to determine what the nature of the image is, how to locate it and obtain permission to use it, and how to find related pieces of information (both analog and digital). Some of the information may not make sense to you if you're not a specialist in British archaeology (e.g. the 622 345 in the 'Coverage' element refers to the United Kingdom's National Grid Reference system) and some information may be superfluous if you're not a geophysicist (e.g. a Geoscan FM-18 gradiometer is a type of instrument used to conduct magnetometry surveys), but the basic information (e.g. this is an image of an Iron Age settlement in Scotland) is all there.

Now imagine if all these metadata entries were interlinked automatically, and your computer could navigate quickly between them. Thus, you might click your mouse button on the 'National Monuments Record of Scotland BW5154' in the 'Relation' element and your web browser would automatically take you to more detailed information about the holdings of the NMRS in Edinburgh. This kind of seamless interconnection isn't yet possible but it will be in the near future, thanks to metadata cataloging.

10. Things to notice in this archaeological example

If you examine the Whitrighill metadata closely, you'll notice several points of interest.

First, all the Dublin Core elements can be repeated as many times as necessary to describe a particular resource. So, for example, the 'Coverage' element is used to refer to five ways of recording the spatial and temporal coverage of the site. Just as elements can be repeated as many times as is necessary, so too can elements be skipped completely if they are not relevant for a particular resource. It's a good habit, however, to be explicit regarding missing elements so users know whether it is because they are not relevant or simply that the values are unknown.

The second thing you might notice is that elements can be broken down into 'Types'. For example, in the 'Author or Creator' element there are two types of creators listed. The first is the person who actually supervised the collection of the geophysics data in the field, and the second is the person who processed the image that you saw above. Both people contributed to the intellectual content of the image which you looked at, and both types of contributions can be recorded in the 'Author or Creator' element.

If you are already familiar with the Dublin Core, you may have noticed that no 'Schemes' were listed. The creators of the Dublin Core very cleverly included a technique for standardising the format of entries. The 'Scheme' modifier allows the person creating a metadata entry to specify if the entry conforms to an international documentation standard. For example, suppose a staff member at the Royal Commission on the Historical Monuments of England (RCHME) were creating a metadata entry for Stonehenge. He or she might draw the 'Subject and Keyword' entry 'Henge Enclosure' from the helpful RCHME Thesaurus of Monument Types (Royal Commission on the Historical Monuments of England 1995). By listing this thesaurus as the 'Scheme' used to record the 'Subject and Keyword' element in Dublin Core, the person creating the metadata would be indicating a valuable tool anyone could use to determine precisely what kind of monument was being referred to.

11. Future challenges

Now, don't let the above example fool you. The adoption of metadata in archaeology is not quite so simple. There are several issues still to be worked out: agreeing a standard metadata framework or frameworks expanded to include all relevant archaeological information, linking Dublin Core elements to schemes in the form of existing archaeological standards, and automating the potentially time-consuming process of metadata generation.

In the United Kingdom, work on a standard metadata framework for archaeology is being led by the Archaeology Data Service (Richards 1996). A workshop series bringing experts together to work on an archaeological adaptation of the Dublin Core began in March 1997, and the report from this project (and similar projects in history, performing arts, text studies, and visual arts) is expected by July 1997. Participants in this workshop began the important process of evaluating all Dublin Core elements to see if each is important for archaeologists (the conclusion was a resounding yes!) and also began listing 'Types' and 'Schemes' which are relevant to the metadata cataloging of archaeological data. Please contact the authors if you would like more information about this project.

A pilot project called 'Accessing Scotland's Past' is also underway in Scotland which will use metadata to link 19,000 site records in the National Monuments Record of Scotland to records about the same sites in regional Sites and Monuments Records. Project partners include the Archaeology Data Service, the Joint Information Systems Committee, the Royal Commission on the Ancient and Historical Monuments of Scotland, the Scottish Cultural Resources Access Network, the Shetland Amenities Trust, and the West of Scotland Archaeology Service.

12. Conclusion

Metadata is a hot international topic now because it seems to be the best strategy available for managing the large quantities of distributed information newly available over the Internet. The Dublin Core metadata system seems particularly useful as it is designed to describe all information at a basic level in a way which both humans and machines can understand. Metadata matters very much to archaeologists as it is essential in archaeology that our data can be easily located and inter-connected with related resources. This is because our digital information is in some ways more important than digital data in other disciplines: it is created from the destruction of primary archaeological evidence and is our only record of that lost primary data source.


© Internet Archaeology
Last updated: Wed Apr 30 1997