e-Framework services for course evaluation
Download the full report
This project provided an e-Framework toolkit to enable users to create applications that provide services and consume services for course evaluation.
Executive Summary
The aim of the project was to provide an e-Framework toolkit to enable users to create applications that provide services and consume services for course evaluation. While such a deliverable was clearly in focus, the project had a second, and possibly more important aim - to explore the JISC e-Framework and its applicability to such service oriented architecture (SOA) e-Learning applications development.
The toolkit was achieved by developing portlets (portal applications, written in Java) within an open source portal framework (Liferay), accessible through a web browser. The portlets attach to reusable services.
The exploration of the applicability of the e-Framework involved re-using services developed in the R2Q2 project for rendering and accepting responses to multiple-choice and free-form questions, and ‘re-using’ the architectural design of portal, portlets, and services from the Collaborative Orthopaedic Research Environment (CORE) project.
Outputs and results
Details for downloading and deploying the system
The toolkit is fully functional (users can author course evaluation items and questionnaires, publish them, respond to them, and see their results) and a demonstrator is also available.
A difference from the original plan, however, is that the results service is simple and basic, and not nearly as sophisticated as originally envisaged. This was the major consequence of the fact that the re-use of services was not as simple and easy as had at first been assumed.
Using the e-Framework and SOA approach during development revealed that the SOA approach may be better suited to larger projects rather than this smaller kind of project. In particular, it took more time to adapt and accommodate the R2Q2 services than originally expected. And, it took more time and effort than originally anticipated to integrate the services within a portal framework. The idea of simply and easily re-using existing services in new application development must be treated with considerable caution.
On the other hand, re-using the architectural design (portal, portlet, and services) from the CORE project undoubtedly allowed a somewhat accelerated schedule, a shortening of the ‘learning curve’, and better project management, than might have been the case with a ‘conventional’ software development approach.
Implications and conclusions
As in any software development, attention must be given to the granularity of the modules or services involved, and attention must be given to the potential for re-use of existing modules or services. While the SOA approach promises, on paper, that re-use is made easier and applications development is accelerated, this project did not yield such clear outcomes in practice. Some of the project team felt that a better toolkit would have resulted if its services had been developed without the requirement to re-use R2Q2 services.
On the other hand, the project management, and the application architecture, certainly benefitted from re-using the SOA architectural design (but not the services) of the CORE project.
A side effect was that the R2Q2 services were considerably improved as a result of their use in this project, to the benefit of following projects such as AsDel, AqRate, and Minibix.