Traditional publication of research projects in the Arts and Humanities often comprises a series of journal papers, a popular summary, an academic monograph, and sometimes an offline research archive. In comparison with other subject disciplines, Arts and Humanities have been relatively slow to embrace electronic publication. There are growing numbers of online databases, but these are rarely linked directly to interpretative analyses. In addition, the Arts and Humanities community has no conceptual framework for evaluating online databases, which results in uneven and unsystematic reward for those who create them. Given that the supporting data are invariably in electronic format, the advent of e-publication allows the combination of different forms of dissemination and the adoption of an integrated approach.
The Linking Electronic Archives and Publications (LEAP) project, funded by the Arts and Humanities Research Council (AHRC) under the ICT Strategy Programme, aimed to explore some model solutions. Archaeology, and the related cultural heritage disciplines, provided an ideal testbed for exploration. The use of ICT is relatively advanced in Archaeology. There is a rich variety of data formats in use, including text and colour images, databases, Geographic Information Systems (GIS), Virtual reality visualisations, video and geophysics.
The LEAP project used the existing infrastructure of the e-journal Internet Archaeology
and of the Archaeology Data Service
to provide four sustainable exemplars of multi-layered e-publications and e-archives, capable of wide implementation across the Arts and Humanities. Internet Archaeology
was established in 1996 with start-up funding from the JISC eLib
programme. It was the first refereed online e-journal in Archaeology and has been very successful in gaining international recognition as a high-quality academic journal. The journal currently has over 70,000 hits per day to its pages, from over 120 countries. It is now in its 23rd volume and attracts a continuous stream of offers of high-quality research papers. Internet Archaeology
is unique in Archaeology in that it is a multimedia journal available exclusively on the Web; it has no print equivalent. It includes elements that would be impossible in a paper publication, such as searchable databases to analyse online; full-colour, interactive images; video footage; virtual reality models and access to related digital archive material. This allows the subscriber to choose the level of detail required through a variety of indexing and searching methods. Internet Archaeology
is currently funded by a mixture of a JISC site licence to provide free access to UK Higher and Further Education, home and overseas institutional and individual subscriptions, plus a small proportion of publication subventions and advertising revenue. The journal is archived by the Archaeology Data Service, also hosted by the University of York. The Archaeology Data Service (ADS) supports research, learning and teaching with high quality and dependable digital resources. It does this by preserving digital data in the long term, and by promoting and disseminating a broad range of data in archaeology. The ADS promotes good practice in the use of digital data in archaeology. It provides technical advice to the research community, and supports the deployment of digital technologies. It receives funding from the Arts and Humanities Research Council, the Natural Environment Research Council, the European Commission, English Heritage, Historic Scotland, the Higher Education Academy, the Heritage Lottery Fund, as well as various agencies responsible for developer-funded archaeology.The Exemplars
In order to demonstrate the extensibility of the project across the Arts and Humanities domains, exemplar articles were chosen that were of thematic or cultural interest to other communities.
The exemplars attempted to provide a novel and imaginative additional form of dissemination for four research projects of high academic quality, and investigate the ways in which e-publications can be interactive, multi-layered and underpinned by supporting data. This was achieved for example, by deep linking to individual elements of the online archive from the interpretative narrative, embedding an online GIS of archive data within the narrative, or making data queries (linking) from within the text to the searchable data in the archive.
Two calls for papers were widely publicised, the first in June 2005, and the second in January 2006, with grants offered to facilitate staff buy-outs or the hiring of research assistants to assist with the preparation of the exemplars. Selection was undertaken by the Internet Archaeology
Editorial Board and ADS Advisory Committee in conjunction with other Arts and Humanities Data Services heads.
Most applications were from archaeology and closely related fields, partially reflecting the higher level of ICT development and awareness within the discipline. Exemplars were selected on the basis of (i) academic significance, (ii) potential of value-added by the ability to drill down from synthesis to primary data, (iii) ability to deliver within the project timescale, and (iv) inter-disciplinarity. Priority was also given to projects that had received AHRB/AHRC funding but which might benefit from an additional means of dissemination. Four projects were selected: ‘Changing Settlements and Landscapes: Medieval Whittlewood, its predecessors and successors’; ‘Joining the Dots: Continuous Survey, Routine Practice and the Interpretation of a Cypriot Landscape’; ‘Silchester Roman Town Insula IX: The Development of an Urban Property c. AD 40-50 - c. AD 250’; and ‘The landscapes of Islamic Merv, Turkmenistan: Where to draw the line?’. Changing Settlements and Landscapes: Medieval Whittlewood, its predecessors and successors
This article presents an interpretative synthesis of the development of a medieval landscape in the English Midlands. Readers are encouraged to explore their own research agenda and to develop different readings of the evidence on which alternative models of medieval settlement and landscape change can be built. The link between the article and the archive was through a Web GIS interface. This allowed users to access and interrogate the spatial data from the archive directly alongside multiple interpretations within the article.
Figure 1: Screenshot from the Whittlewood article highlighting the Web GIS interface inline with the article text. The pop-up window in the bottom left of the image shows a results page from querying the Web GIS. This includes attribute information related to the feature and a link to archive material (a comma separated value table in this case)
Joining the Dots: Continuous Survey, Routine Practice and the Interpretation of a Cypriot Landscape
One of the major challenges facing intensive surface survey is how to interpret surface artefact scatters in terms of past human activities and relationships. This study uses web-based GIS and database technologies to provide a complete landscape data set and a fully integrated interpretative text carefully grounded in current landscape theory. The material comes from the Troodos Archaeological and Environmental Survey Project (TAESP), which carried out intensive survey in the northern foothills of the Troodos Mountains in central Cyprus between 2000 and 2004.
Like the Whittlewood article, TAESP used a Web GIS interface with spatial data served from an ArcIMS server hosted by the ADS. Users could also interrogate the project’s database from within the article.
Silchester Roman Town Insula IX: The Development of an Urban Property c. AD 40-50 - c. AD 250
This article discussed the development of an urban property in the Roman town of Calleva Atrebatum (Silchester) from the late 1st to the mid-3rd century AD. Three successive periods of building with their associated finds of artefacts and biological remains were described and interpreted with provisional reconstructions of the buildings. Links were provided to a copy of the Integrated Archaeological Database (IADB), archived by the ADS, holding the primary excavation and finds records.
Figure 2: A reconstruction of Timber Buildings 1,2, & 3 from Silchester (drawing by Margaret Mathews)
The landscapes of Islamic Merv, Turkmenistan: Where to draw the line? (Forthcoming, Internet Archaeology issue 25)
Merv straddles one the main branches of the ancient Silk Roads that connected Europe and Africa to the Far East. The succession of cities date from the 5th century BC to the present day. The article develops and tests the methodology of documenting interpretation (and uncertainty) based on a variety of data sets. The publication will have a Web GIS interface similar to the Whittlewood and TAESP projects as well as video and audio content, and sections of the text will link directly, for example, to images held by the ADS.
The LEAP project raised a number of questions due to the novel nature of dissemination, both from the publication and the archive standpoint. Many of these questions inspired useful and productive discussion on digital dissemination in general which went beyond Internet Archaeology and ADS procedures and which will provide feedback into future updates to those procedures.
One issue related to how the resulting resource should be assessed and peer-reviewed. Since we were deliberately blurring the boundaries between the electronic archive and the publication, should the peer-reviewer also comment on the archive data to the same level of detail as they would the narrative 'text' and supplementary material? It is common Internet Archaeology practice to make all supplementary materials available for the referee. But when this also constituted the data deposited with the archive,, then this makes a truly full assessment a much larger undertaking by the selected referee.
Another unforeseen outcome of the LEAP project was the question of how archives should be structured so that they could be linked with minimal intervention. This becomes a very serious issue if the e-publication and archive do not share a common infrastructure. Furthermore, it was important to manage author expectations. Those who were used to Desktop GIS, for example, found it difficult to accept the reduced capabilities of Web GIS. Some early design visions clashed with accessibility and usability requirements so robust management was necessary.
Some more general challenges arose. The LEAP project worked with the intention that the narrative published in Internet Archaeology should tell a specific story that is dependent upon links to a data archive. But we realise that in doing so, we run the risk of forcing one particular interpretation upon the reader with the possibility that further research on the data may actually be harder to carry out. There is a fine line between supplementing the data and informing the reader and controlling the data they can look at.
An issue of citation highlighted by the LEAP project was the extent to which intellectual credit could or should be given to all authors of the integrated resource. Now that the archive is essentially part of the publication, does everyone involved in the creation of it, the databases or GIS, deserve to be listed as an author? If this were the case there could be dozens of authors due to the way archaeological projects are frequently conducted.
An even more difficult question was how queries or views into the archive data should be cited, or even if they should be cited at all. This situation is the product of the deliberate ambiguity of where the publication ends and the archive begins. If the archive should be cited in a similar manner to a journal article, then this could be taken to its logical conclusion that any query run against a database should be cited. In reality, the publication is merely linking to an archive, so while the interface into the data archive may be embedded in the publication, it is still simply an interface. As such, it was concluded that citation of the archive as a whole should be made when referencing data from a linked electronic publication and where a researcher identifies something in the data that has not been mentioned in the interpretation, then both the archive and the publication should be cited.
A significant concern of the LEAP project that cannot be answered immediately is the question of interface sustainability. Information and communication technologies develop at breakneck speed. Left in that wake are the web developers and designers trying to keep their interfaces and applications in a functional state. This is both a short term and long term battle, although the long term questions are of the most concern. There is a desire to keep interfaces consistent throughout time to avoid confusion on the part of the user. This becomes a very daunting and unpredictable task due to the nature of the Web and its associated technologies.
Exacerbating the potential problems are the uncommon and bespoke interfaces necessary for Web GIS and databases. While web browser developers endeavour to maintain backwards compatibility, this is not guaranteed; nor should it be assumed. As browsers evolve, maintaining these interfaces could become a massive undertaking. The ADS takes the approach of preserving the raw, underlying data rather than the interfaces through emulation. With the data preserved and migrated as formats evolve, the ADS can build new interfaces to present the material as necessary. It will be important that these new interfaces have a look and feel similar to the original interface to ensure continuity.
The ADS and Internet Archaeology share a common technical infrastructure and both currently create their Web GIS interfaces with ArcIMS and their database interfaces in Coldfusion. Both of these are proprietary solutions which have potential migration issues as they become obsolete. This will not affect the raw data but will potentially create more work for future editors/data curators as they are required to develop new interfaces for the data. The future migration of LEAP interfaces remains the biggest unknown for this project, and one which we will be unable to answer until that time comes. It might be that the migration is a simple 30 minute job 10 years from now or a 10 day job 30 years from now, but we will not know until we get there.
Finally, before future LEAP-style publications can be produced, a shift in the way archaeologists prepare, create and think about data must occur. Traditionally, the archive has been something of an afterthought, pieced together after the fieldwork is finished. If the archive becomes part of the publication, however, more care from the outset of an archaeological project must be taken with regards to data creation. This means that full metadata and documentation of the data must be created. Putting the raw data alongside the publication may force data producers to take more care in creating their data. Archaeologists can sometimes create data knowing that few outside the project will ever see it. This new exposure will hopefully force archaeologists to break the uneven data management habits that have historically afflicted the discipline.
Making the raw data more accessible also makes the author more open to scrutiny. This may not always be welcome. While one of the goals of the LEAP project was to allow users to draw alternative conclusions, scholars have been historically protective and defensive of their data. We hope that as more archives are linked in with publications, scholars will themselves become more open to allowing other to interrogate and re-interpret their data. To accomplish this will take more than a few exemplars, but rather a change in the way researchers think about their data.
The LEAP exemplars have already had significant impact in the ongoing debate about the development of electronic publication. The number of people directly involved in the project through the four exemplars has had an immediate impact on the dissemination of good practice. Individual publications and archives have each seen high levels of usage, although it is too early to analyse general trends. Nonetheless, the web access statistics show high levels of usage for each exemplar from the outset, including individual archive files being downloaded an average of over 100 times per quarter.
Julian D. Richards, firstname.lastname@example.org
Judith Winters, email@example.com
Michael Charno, firstname.lastname@example.org
University of York