Together with partners from various academic institutions, the Archaeology Data Service (the UK partner in the ARENA project) has recently been involved in a Joint Informations Systems Committee (JISC) funded project: the Contextual Resource Evaluation Environment (CREE). Coordinated primarily by the University of Hull, CREE aimed to investigate how existing search systems can be realised as portlets within an academic/institutional portal environment.
It was decided that the HEIRPORT portal was an ideal candidate for the CREE investigations. CREE coincided with a research programme that had already been initiated within the ADS, and it allowed a real life case study for CREE of a standards-based search tool that could be adapted. HEIRPORT is managed on behalf of the Historic Environment Information Resources Network consortium (HEIRNET) by the ADS (Fernie 2003). By combining the flexibility of the Z39.50 protocol together with web technologies such as XML/XSL, HEIRPORT demonstrates how information from different sources pertaining to the historic environment can be gathered together and presented in one place in the same way as the ARENA portal does (Austin et al. 2002).
While the technologies and application framework that underpin HEIRPORT are relatively mature, the rapid development of current web standards means that it requires a constant effort to maintain the quality of service and ensure technical evolution. Launched in 2002 it has already been subject to one major rebuild (Kilbride 2003) and a parallel phase of international development (Kilbride 2004). Both of these more recent initiatives imitated simplistic portal personalisation by offering distinct, pre-defined but user selected views of the services available: one phase offered six different simultaneous interfaces that coincided with stereotypical user behaviours, while another phase saw the implementation of interfaces to support different languages. Though progressive, neither of these offered personalisation in any strict sense. Many of the emerging aspects of online portal frameworks were either absent or only available in principle. These include self-adapting personalisation of content, but also extend to the recycling of generated information and support for web services. These features have become more important to the definition of a web portal in the last few years (Miller 2003), so the term 'portal' could have become somewhat misleading when referring to the HEIRPORT search system. Thus, engineering elements of the functionality of HEIRPORT within a more broadly defined institutional/academic portal environment would be advantageous both for information providers, such as the HEIRNET partners, as well as for the end-users. The same can be said for the ARENA portal, which could also be recreated to plug into other institutional or academic portals.
The biggest factor influencing the portlet design was understanding the existing search system. Working with an existing framework meant that development needed to be cognisant of the existing tools and not prejudicial to the quality of service. This aspect of portlet design is relevant to any portlet developer presented with an existing or potentially unknown system to adapt as a portlet. Development planning had to incorporate an understanding of the existing system and be congruent with existing services before work on the portlet could begin in earnest.
Rather than re-writing the HEIRPORT portal from scratch as a portlet application, it was decided at an early stage that the portlet would act as a portable interface/wrapper to the existing HEIRPORT client. Specifically, the portlet would provide various mechanisms to: create a query, send the query to the HEIRPORT server, receive result sets as a response from the server as plain XML, and, fundamentally, provide a mechanism to manipulate the XML data in a way best suited to a portal environment.
Although implementing the above mechanisms would satisfy the requirements of the CREE investigations, it was also decided to include 'value-added' functionality, perhaps making use of specific portal features. It was hoped that this would enhance the overall HEIRPORT experience and provide users with functions that have become commonplace in other websites.
The following features were thus planned for the HEIRPORT lite portlet:
© Internet Archaeology/Author(s)
University of York legal statements | Terms and Conditions
| File last updated: Tue Sep 27 2005