JISC

Start date: 1 February 2005

End date: 31 January 2007

Funding programme: Virtual Research Environments programme (phase 1)

Project website: http://www.grids.ac.uk/Sakai

JISC theme(s): e-Research

In this project we will extend the Sakai open source and open standards portal framework. It will host a number of integrated tools and robust distributed services and become the principal user interface for a fully-functional Virtual Research Environment (VRE).

To accomplish this, we will seek widespread community input to expose existing tools and services as Java portlets, SOAP-based or peer-to-peer Web services for use within the VRE. Using this approach, the project will create a long-lasting, maintainable and extensible VRE which will provide both reliable collaboration tools, together with a powerful distributed research toolset. Some of the VRE tools and services we want to expose are UK adaptations of those directly available from Sakai and OGCE, the US NMI portal. Others are to be achieved in association with a UK community of VRE users.  

Aims and Objectives  

This project aims to address the requirement for a single point of access to a comprehensive set of Grid and collaboration services in a VRE.  

To achieve this we have identified, in extensive discussions with the US developers, four areas of framework extension for which the partner sites will take individual reponsibility:

• Identification, or specification, of an XML grammar for describing collaborative research (Lancaster);
• A Service Authentication and Authorisation System (Oxford);
• A JSF-based Web service interface generator (Daresbury);
• Interface to Peer-to-Peer (P2P) Tools (Portsmouth).

These extensions will permit the integration of a wide range of tools, some as part of the Sakai application (using the Java standard JSR-168) and others interfacing to remote services (using a JSF Web service interface or XML standard for remote portlets WSRP).  

The main benefit of this VRE will be the ease of access to powerful distributed Grid and collaboration tools consistently from one interface. The success of this strategy will be determined by the VRE's level of use. The ease of deployment and integration of tools will also be assessed.  

Three metrics are proposed to help evaluate the success of the VRE for end users. The first is a single user metric and involves basic log analysis to determine the number of hits per day, measured over the final month of the project. The second involves counting the number of user created worksites and applying a weighting based on the number of users added to the worksite. The third is a collaboration metric and will involve determining the average number of simultaneous users in a Sakai worksite, again averaged out over the final month.  

Project Methodology  

To achieve the goals of the project the work is split into four work packages on which the partners will work together:

• WP1: Web Services and VRE Extensions to Sakai Framework;
• WP2: Integration of UK VRE Tools into Sakai;
• WP3: A Robust Sakai Demonstrator and Services;
• WP4: Community, Documentation, Training and Awareness.

Individual tools which are being documented and ported will be installed in working Sakai portals at one or more of the sites, and as they are tested will be available on-line for demonstration purposes from the main project portal at Lancaster. Documentation and related information will be provided via this portal and the project Web site. 

Deliverables  

The principal deliverables of the project will be: (1) intimate knowledge of the Sakai project and its tool integration philosophy; (2) end-user community experiences of collaborating and accessing Grid, information and e-Learning resources from 2nd Generation Web portal; (3) a rich set of tools and framework extensions which will be provided to the Sakai developers group for subsequent uptake in other projects. 

The list of proposed tools includes: 

Search Tools: Context-based Information Retrieval, Collection Cross Search, Google Web service interface, Application Discovery, Scientific Data Query, UDDI.  tools as appropriate. 

Collaboration/ Discussion: IRC (Internet Relay Chat), Blog, Research Scheduler, Video Conferencing, Distributed Whiteboard, Distributed Display. 

Anaysis Tools: Scientific Calculator, Active Spreadsheet, Statistical Computing, Data Management, Simple Visualisation, Grid Information, Network Information, Grid computing. 

Publication/ Documentation: Word Processor, Distributed Presentation Tool, Bibliography Builder, XHTML Editor, LaTeX Processor, Simple Interactive Plotting, Project Publication, Application Publishing. 

Others: Portal Statistics, Personal Information Manager, Shopping Cart, Bugzilla, CVS, Authorisation Policy Management. 

Training and Awareness: Research Assistant and Personal Information Guide (PIG), CopperCore Tool, Tool Templates, Documentation Manager, Helpdesk interface. 

These will be prioritised with help from the project partners. 

Stakeholders  

Stakeholders in the project include:

• Participating institutions and projects - will gain exposure to the capabilities of 2nd Generation portals for e-Research and e-Learning; 
• Developers - will learn new skills in Java portlet and Web services technologies;
• UK e-Research community - will be provided with a rich set of tools accessing distributed services for uptake in support of their work;
• JISC - will be provided with feedback on the Sakai portal as a delivery vehicle for the services comprising a Virtual Research Environment and on the commonality of these services with those of e-Learning and the Information Environment;
• Sakai developer's worldwide - will have access to the framework extensions and tools we develop if they are put into open source.

Evaluation Plan  

We will analyse a combination of individual logs, structured interviews, observation of work processes (and changes to processes) and usage statistics to assess the suitability of the Sakai software for the TLRP community. 

A semi-quantitative research productivity model based on Wenger's Communities of Practice work will be developed and changes in activity recorded along the dimensions that determine research productivity. This work will use qualitative analysis software such at ATLAS-T and NVivo.  

We expect to be able to demonstrate increased collaboration and hope to be able to demonstrate increased productivity. A modest amount of time will be devoted to classic usability testing of the Sakai software with the TLRP researchers. 

Project Partners  

Partners will be included from interested e-Science projects, such as the National Grid Service, e-HTPX, e-Minerals, MyGrid, Integrative Biology, and the National Centre for e-Social Science.