"Beyond GIS": Geospatial resources and services for scholars in the humanities
The study and use of geospatial information is in itself a discipline, with its own literature, journals, conferences and research projects. There are well-known crossovers between this discipline and humanities subjects, which has led to the development of (usually) small specialized communities within different disciplines. These usually comprise of users of Geographical Information Systems (GIS) technologies and, in many areas, have become well-established. ‘Archaeological GIS’ for example, and ‘historical GIS’ are now widely recognized tags for work done using GIS in archaeology and history, a fact reflected in the strong geospatial component regularly present in domain-wide conferences such as Computer Applications and Quantitative Methods in Archaeology. However, in most applications, GIS is a set of tools which enables relatively straightforward visualization functions of geospatial information on the desktop, and/or permits a number of fairly familiar analytical procedures on suitably prepared and formatted data. Until recently, most GIS software was proprietary, expensive, and not generally accessibly to the non-specialist. Although the growth of Open Source GIS has certainly opened up the developmental side of the technology, it now exists within an intellectual context far broader than anything that the label ‘GIS’ was ever intended to refer to. For this reason, I have followed the Open Geospatial Consortium convention of using the more inclusive term ‘geospatial’ as a generalist term of the tools and technologies discussed.
In recent years, there has emerged a huge range of services, resources and applications for producing, accessing, using and managing geospatial materials. These have a wide range of functions, which include (for example) ‘mashing up’ different services, delivering datasets, and standardizing information. However, all may be seen, in one way or another, to be responses by communities who have long worked with geospatial information to the possibilities engendered by bigger, faster and more reliable computational networks. This effect can be seen very strongly in the world of e-Social Science, and in the emergence of the so-called ‘Geospatial Web’, a concept which applies the data sharing principles of the internet – whose main and most familiar function is as a vast collection of interlinked pages containing textual, image and video data – to deep and complex bodies of geospatial information. The emergence of these networks has also bought about new ways of working, as documented (for example) by the AHDS e-Science Scoping Survey. And while social networking software and other so-called Web 2.0 technologies have not necessarily revolutionized research in the way some predicted , the capacity provided by these technologies for users to create, add and annotate their own geospatial and cartographic content has nonetheless bought major benefits to the field (some practitioners sometimes refer to this as ‘neogeography’).
Many of these responses are directly relevant to the humanities, and reflect some of the A&H e-Science Initiative’s innovative approaches to complex and/or diffuse data (an excellent example of this in the area of geospatial data is Birmingham University’s Medieval Warfare on the Grid: The Case of Manzikert project ). This paper briefly reviews some of the main practical developments and theoretical issues, and seeks to provide a practical introduction to some important services and resources. Much of it has been developed from the discussions at a workshop supported by the AHRC ICT Methods Network, Space and Time: Methods in Geospatial Computing for Mapping the Past, held in July 2007, although extra material has been added. The aim is explicitly for breadth rather than depth – individual resources are discussed briefly, but in each case links and references to further information are provided. Even so, the field is very large and constantly expanding, and there will inevitably be geospatial topics, resources or services that are relevant to the humanities but are not discussed here.
Associating information with geography
In a recent review of the area, Martyn Jessop notes that ‘[t]here is a requirement for more central archives and a metadata schema that would allow [humanities data] resources to be discovered and shared across archives’ . This reflects a broadly-held belief that geospatial information offers a suite of very powerful methods for linking humanities resources from disparate sources: most kinds of data can be associated in some way with geographic information that can in turn constitute metadata. E.g. a photograph can be associated with a specific point on the earth’s surface by a caption specifying place, coordinates from a GPS-enabled camera, reference in a text, and so on. Texts can refer to specific places or types of places in a great variety of ways. Finds on archaeological sites are often recorded using theodolite readings, or other spatial location techniques (which produce site-specific datasets that need to be related to a global coordinate system; or imagery with spatial distortions caused by perspective that need correcting; a process known as orthorectification). Many of the possibilities for linking and cross-searching datasets using geographical information were explored in an AHRC-funded workshop in the Arts and Humanities e-Science Initiative, Geographical Information Systems e-Science: Developing a Roadmap. This expressed a theme of ‘Spatio-temporal e-science’, by which information is linked using placename information, and deployed to the research community using the data grid.
Associating information which is not geospatial with a geospatial framework or system (like latitude/longitude) in this way is known as georeferencing. This is fundamental to how geographic information is stored and retrieved. Georeferencing can be either formal, i.e. based on a mathematical string expressed in degrees, or some other decimal numeric projection; or informal, which usually means based on place names, or some other textual or semantic description. A comprehensive overview of georeferencing, and bibliography, may be found in Hill 2006 .
Standards, web services and file formats
Possibly the best best-known and widely used geospatial resource on the web is Google Earth; figure 1. Many of the basic functionalities required by humanists are available free of charge. Google Earth offers global coverage of satellite imagery and the facility of overlaying Google Maps files. The chief advantage of Google Earth is that it can be easily and intuitively navigated, either by a ‘joystick’, which appears in the top-right hand portion of the screen, or by searching place names. This function also supports basic, yet intuitive, semantic searching, e.g. ‘hotels near San Diego’, or by postcode. Most importantly however, Google Earth supports the creation and viewing of files in Keyhole Markup Language (KML), an XML-based standard for encoding spatial information (Google Earth’s original name was Keyhole). This allows the definition of features such as placemarks, shapes, 3D models, descriptions using texts (in much the same way that ‘conventional’ GIS platforms operate around the concepts of point, line and polygon). Each KML object is associated with earth’s surface using latitude and longitude coordinates. KML’s chief benefit is that it is extremely flexible and easy to use, and interoperates with other earth browsers. Other functionalities, such as TimeStamp, which references a particular moment in time for any given feature, and TimeSpan, which specifies a beginning and an end for a feature that changes over time, are available on Google earth’s premium service.
[img_assist|nid=1477|title=|desc=Google Earth Image|link=node|align=left|width=415|height=349]
The flexibility of KML’s XML basis makes it an ideal format for ‘mashing up’ geospatial data with different media. One example comes from the ‘Ogle Earth’ blog, where photos taken with a georeferenced digital camera, uploaded to the photo sharing site Flickr, and overlaid onto the Google Earth surface image (figure 2). This kind of approach has obvious potential benefits for the way images in humanities research projects are related to maps, and any other type of geospatial object.
[img_assist|nid=1480|title=|desc=Figure 2: Overlay of Google Earth base data using Flickr|link=node|align=left|width=249|height=198]
The Open Geospatial Consortium (OGC) is a grouping of over 300 organizations from industry, HE and governmental agencies, whose aim is to ‘[manage] a global consensus process that results in approved interface and encoding specifications that enable interoperability among and between diverse geospatial data stores, services, and applications’. JISC’s relationship with the OGC is based primarily in the GRID OGC Collision programme, which comprises the SEE-GEO project at "Edina, which is concerned with secure access and management of geospatial data, and SAW-GEO, which is developing workflows and methodologies for orchestrating web services for OCG resources.
Resources, services and institutions
There are several well-known projects which act as geographic information providers, and are geared towards reasonably generalist scholarly audiences. The Electronic Cultural Atlas Initiative (figure 3), based at the University of California Berkeley, is one such project. ECAI defines itself a ‘cultural atlas’ as ‘a matrix of spatial and temporal information with associated cultural information or attributes. It must include at least one dynamic map with multiple layers, or multiple maps.’ The project adopts a federation approach, whereby appropriate individuals and organizations. These include data and/or service providers, scholars, archivists and other information professionals, who sign up as ‘affiliates’. A wide range of resources are available through the ECAI interface, which can be searched by placename, or via Google Earth (which underlines the importance of the point made above about the flexibility and adaptability of Google Earth, and the ease of interoperating it with other programmes).
[img_assist|nid=1479|title=|desc=Figure 3 Electronic Cultural Atlas Initiative|link=node|align=left|width=419|height=281]
Another well-known application closely related to ECAI is the TimeMapTM project, run by the Archaeological Computing Laboratory at the University of Sydney. As the name suggests, the aim of TimeMap is to move beyond historical map data being presented into simple time-stamped GIS ‘snapshots of a former present’, and instead to meaningfully integrate time. This is done by providing a metadata clearing house for geospatial material, indexing any geodatabase stored on the web, and providing a Java client (TMJava) which renders time-based geospatial media animations in Flash. It is extremely useful if a researcher wants straightforward yet high-quality
The Alexandria Digital Library based at the University of California Santa Barbara is an example of a very large ‘geolibrary’. The ADL employs a metadata schema which formally describes a wide range of geospatial information using controlled vocabularies. A Feature Type Thesaurus (FTT) allows the user to drill down from very broad categorizations at a regional level to much more detailed and refined level, and a gazetteer provides a homogenous and consistent reference for placenames associated with the FTT. This A full technical outline of the ADL is available in Hill 2006.
The Archaeology Data Service (ADS) is an important provider and manager of geospatial information. Its ArchSearch facility allows users to search the ADS collections - an array of archaeological datasets containing over 1 million records - via an interactive map of the British Isles, by placename or by Ordnance Survey Grid Reference. The ADS also contributes to several research and development activities relevant to geospatial computing in the humanities, including for example the ARENA project.
Edina, based at the University of Edinburgh, is a major JISC-funded provider of digital geospatial resources for HE. Most significantly these include Digimap, which allows registered tertiary education users to download digital tiles of the UK map data for the UK, or to access them using standard GIS software packages. Another important Edina service is UKBORDERS, which provides GIS-formatted details of UK administrative boundary data, which has a wide range of possible uses in historical GIS and census research.
[img_assist|nid=1478|title=|desc=Figure 4: Landsat 7 data tiles for download from the MIMAS Landmap service|link=node|align=left|width=241|height=249]
Also JISC-funded MIMAS, at the University of Manchester, provides the Landmap service (figure 4). Landmap provides high-quality satellite data free to licenced HE institutions. For some applications the imagery available via Google Earth may not be considered ‘research quality’; Landmap data in most cases will be.
Online mapping communities
As mentioned in the introduction, the emergence of so-called ‘neogeography’ and Web 2.0 tools and methods in the geospatial humanities been accompanied by the rise of numerous online communities, fora and discussion groups on every aspect of the subject. Such groups are often excellent platforms for getting advice, contacting experts in particular areas, or disseminating research or raising issues. These include email lists, interest groups on social networking sites such as Facebook (e.g. GISuser), wiki-based groups, and other familiar forms of relatively informal, internet-based collaboration. As with all areas of specialism, some online activities in this area there are more reputable than others, and there are far too many such groups to list individually. However, three established clusters which can be trusted as sources of information are introduced here to indicate the kind of activities currently underway.
The Pleiades project, based at the University of North Carolina, is an Open Source based community platform which hosts the data which formed the Barrington Atlas of the Greek and Roman World, a major established reference work for researchers of classical antiquity. The project allows any visitor to the website to suggest updates, or add content to, any aspect of the geographic data (coordinates, descriptive texts, placenames etc) making up the dataset. These are, however, subject to an academic review process, so Pleiades is essentially an explicit hybrid of the open-content Wikipedia model and ‘traditional’ academic publication. This project is a clear expression of the indivisibility of geographic information and humanities data in general, cited above: given that the geographic framework of the Mediterranean region is completely ubiquitous in classical studies, Pleiades’ content and collaborative processes are likely to be of interest to any researcher engaged with any aspect of the classical past.
The Antiquist group is a discussion forum which communicates primarily via a Googlemail list (although it also has a wiki). It has a broad membership including museum professionals, archaeologists, historians and fieldworkers. There are many members of the Antiquist group who have particular interests in GIS and other geospatial tools and methods; and there is a strong technical slant to many of the list’s threads. It is a particularly appropriate forum for any researcher interested in software development of geospatial tools. In 2007, Antiquist moved its wiki to the Arts-humanities.net platform. Arts-humanities.net is the community portal of the AHRC ICT Methods Network, and hosts discussions and publicizes activities on a very wide range of digital humanities topics. This includes geospatial research themes. Topics on the site include the ‘Mapping the Past’, a group currently with 101 subscribers.
Training and research support
As with many other areas of the digital humanities, a lack of generally available training programmes for geospatial computing is a major problem. There are excellent postgraduate courses available of course, but the notion advanced in this paper is that geospatial computing is of practical use to a research constituency far broader than the expert community produced by such courses.
In his paper, Jessop identifies institutional and research cultures as major barriers to the widespread use of geospatial resources in the arts and humanities. He proposes a ‘one stop’ resource to provide training, access and guidance for researchers. In the absence of such a resource, and given that the vast majority of university computing services are not tasked or funded to provide support or training in the use of geospatial software tools, most researchers using those tools rely to a great extent on the online communities referred to in the previous section. Some ‘one stop shop’ functions are provided by arts-humanities.net. Also, those interested in locating training resources may consult the ICTGuides portal, hosted by the Centre for e-Research at King’s College London, which provides advice and listings for relevant training materials and resources.
References and further information
 Units such as UCL’s Centre for Advanced Spatial Analysis have emerged with particular interests in rich geospatial data, and its application to the huge datasets employed by e-social scientists; and the potential of Web 2.0 technologies in so doing.
 e.g. Scharl, A. and Tochtermann (eds) 2007: The Geospatial Web: How geobrowsers, social software and the Web 2.0 are shaping the network society. Springer-Verlag, London.
 P. Anderson 2007: What is Web 2.0? Ideas, technologies and implications for education. JISC TechWatch report TSW0701: http://www.jisc.ac.uk/whatwedo/services/services_techwatch/t...
 M. Jessop 2007: The Inhibition of Geographical Information in Digital Humanities Scholarship. Literary and Linguistic Computing (Advanced Access, 20 November 2007: http://llc.oxfordjournals.org/cgi/content/abstract/fqm041v1)
 L. L. Hill 2006: Georeferencing: The Geographic Associations of Information. Cambridge, MA/London: The MIT Press.
[img_assist|nid=474|title=|desc=|link=url|url=http://www.ahessc.ac.uk|align=left|width=78|height=78]This briefing paper was written for AHeSSC, the Arts and Humanities e-Science Support Centre. It is published here with permission from AHeSSC.