Aims, Objectives and Final Outputs of the project
The overall objectives of the RISE project are to:
- Establish a process to collect and analyse attention data about the use of electronic resources from the EZProxy proxy referral system.
- Create a recommender service using attention data to answer questions such as ‘people on my course are looking at these resources’
- Identify metrics to detect changes in user behaviour as a result of service use.
- RISE will create a personal recommendations service, MyRecommendations for OU users of the EBSCO Discovery Solution (EDS).
- It will explore issues (of anonymity, privacy, licensing and data format/ standards) around making this data available openly and will aim to release it openly so it can be re-used by the wider community in innovative ways.
- RISE will use the EDS API to create a Google Gadget for the OU Google Apps environment and will aim to test in the OU Moodle Virtual Learning Environment (VLE) using features developed by the JISC DOULS project.
- RISE will evaluate the pros and cons of providing recommender data to students of an e-resource discovery service.
- Overall RISE will provide the wider community with an example of the benefits to users of discovery solutions of using e-resource activity data, will aim to make that data available to the wider community, and will provide a tool that can be adapted and reused.
Project outputs will include:
- The release of database schema, documentation, algorithms and code for a recommendations service using EZProxy data as Open Source through a code website such as SourceForge or Google Code.
- Open release of data with a Creative Commons Universal CC0 license (where this is possible).
- Creation and open release of a Google Gadget using the EBSCO Discovery Solution API
- A series of reports including a report on the issues of releasing the data openly, a report on the feasibility of using recommendations within a gadget and an evaluation report of the benefits of using the tools that have been developed.
- This blog will be used to provide updates on project activities. Other dissemination activities will also be explored and used to provide updates on the project work.
Risk Analysis and Success Plan
A risk assessment table is shown below:
Risk | Probability | Impact | Mitigation |
Lack of availability of key staff | Low | High | Plan work for this project into staff workplans. Ensure staffing time is realistic |
Loss of key staff | Low | High | Ensure documentation is kept up to date and that knowledge is shared across the Systems Development team |
Unexpected technical difficulties encountered | Medium | Medium | Work is carefully scoped to ensure it is realistic |
Unable to release search results data owing to reluctance in organisation | Medium | Medium | Business case approved by Library Leadership team prior to bid submission. RISE will engage with other projects and with key stakeholders |
Unable to test Google Gadget in OU VLE environment | Medium | Low | Gadget will be tested in a range of environments both OU-controlled and external |
Lack of engagement from stakeholders | Medium | Low | The Project Manager will draw up a Communications Plan to engage with stakeholders to ensure that they are aware of the project and their needs are met. |
The success of the project will be measured in several ways: the response of users, the take-up of the tools; and the feedback from the HE community.
How measured | What success looks like | |
User response | Survey and informal feedback from students and academics. Analytics data. | Majority of users agree that recommendations are useful and enhanced their use of the search system. Analytics shows positive impact. |
Take-up of tools and data | Usage of tools and data, downloads of tools and data. | Tools are being downloaded several times a week and there are some comments about the tools. |
Community feedback | Feedback. | Wider discussions with community about potential of tools & ways to use the data. |
Intellectual Property Rights
The code developed as part of the project will be released as Open Source through a Google Code (or SourceForge) site. This will be supported through a forum which will be monitored by project staff and by Library Systems Development staff to ensure the sustainability of support for the code. The intention is that any data released as part of the project will be made available as CC0 Creative Commons Universal. Advice will be sought from the Lucero project to ensure that practices are consistent. OSSWatch will be approached for comment to identify appropriate code licence.
All project outputs will be made available free at the point of use to the UK and international academic communities. If monitoring of project output downloads indicates high levels of activity the project will attempt to ensure that project outputs are available through multiple channels to avoid excessive load on individual services. The project will aim to provide some support to adopters but will aim to try to engage with wider communities of interest.
Project Team Relationships and End User Engagement
Members of the Project Team and responsibilities
- Richard Nurse – Project Director. Responsible for ensuring the overall direction of the project and the fulfillment of project objectives;
- Elizabeth Mallett – Project Manager. Responsible for ensuring that the project is managed effectively and delivers objectives, outputs and reporting to target and budget;
- Paul Grand – Developer. Responsible for developing the RISE MyRecommendations system, the anonymisation processes and Google Gadget developments.
- Hassan Sheikh – Technical Consultant. Responsible for providing techical guidance on library technical environment including EDS.
- James McNulty – Technical Consultant. Responsible for providing technical guidance on library technical environment including EZPproxy.
- The Project Team will work closely with other library staff from the Learning and Teaching and Marketing teams to engage with users
RISE will be engaging with end users through a number of mechanisms. We are planning focus groups with students to assess student needs, will provide students with feedback and support mechanisms to gauge the value of the tools to students, and will be carrying out small-scale usability testing. Within a six month project there are practical limitations around the amount of engagement that is possible.
Projected Timeline, Workplan & Overall Project Methodology
RISE comprises four main activities: creating the recommendation system; exposing e-resource search data; creating a Google Gadget; and evaluation.
Workpackage 1: Project Management
This covers the project start-up, including set up of internal and external reporting and project management arrangements, creating the detailed project plan, setting up the project website/blog. It will include day-to-day management of the project. There will be a project wind-down stage ensuring that developments are appropriately documented and processes are in place for on-going support and sustainability.
Workpackage 2: MyRecommendations service creation and implementation
A short period of user requirements gathering will take place to look at the appropriate data to be used and how the database will be structured. A mySQL database will be setup to take data from the archived EZProxy log files and other OU systems to record details such as: who is searching, what course they are on, what did they search for, and what resources have been looked at and when.
RISE will then build a service ‘MyRecommendations’ that allows users to login to a personalised webpage that contains a search box to search the EDS system, feeds back data to them of their activity and provides recommendations based on the activity of other users. For example it can show their recent searches, what other people on their course have searched for, and what popular keywords are being used to search for electronic resources. The system will also allow rating of resources. We will also explore the feasibility of linking to the MyReferences tool developed by the JISC-funded TELSTAR project to allow users to record their references. We plan to consult with other interested parties working in this area to identify suitable algorithms and processes. The new service will be trailed through the library website as an alternative to the standard EDS interface.
Workpackage 3: Opening up recommender data for electronic resources
RISE will aim to make the search data available openly. To do so it will carry out an investigation of the issues, such as privacy, data ownership, licensing, data formats and anonymisation, involved in releasing the data. The project will liaise with the JISC-funded LUCERO project to draw on their experience of releasing linked data through http://data.open.ac.uk and for guidance on the use of course codes and licensing. The project will follow the guidelines set out in the MOSAIC project for anonymising data. The project will look at options to anonymise the data such as hash values or affinity strings.
Workpackage 4: Development of a Google Gadget e-resource search and recomendations tool
A Google Gadget search tool will be created using the EDS API. We will explore and test how recommendations can be provided to users in this format. We will then make the gadget available openly via the OU library website and to the OU Google Apps environment. We will also work with the JISC DOULS which is exploring integration between Google Apps and the VLE. If possible we will test the gadget within the VLE.
Workpackage 5: Evaluation
RISE will carry out a short evaluation to investigate the differences between search, discovery and downloading behaviour of users and non-users of the developments. This will use both web analytics and surveying. A small-scale usability test will also be undertaken.
Workpackage 6: Dissemination and engagement with programme
The project will undertake a programme of dissemination activities to engage with the community. These will include internal staff presentations within the OU Library and across the OU, wider activities such as a Library Seminar, and will look to promote the work at appropriate sector events and channels. Dr Tony Hirst who blogs at http://OUseful.info, is involved with RISE and likely to blog about the project. RISE intends to use regular blog posts, on both the project blog and via the library news blog, to provide updates for the work and many of the project team members are active users of twitter so will promote the project using #OUrise as well as #inf11 and #jiscad. The project looks forward to engaging with JISC programme events and is experienced at doing so with previous JISC projects. Time for engaging with the Programme and Synthesis project has been built into the staff time for RISE.
Project Management Arrangements and Governance
RISE will be managed using the standard OU Project Management processes which are to a modified PRINCE 2 methodology. These processes include the standard project documentation, risks and issues management processes. An experienced project manager – Elizabeth Mallett – has been assigned to manage the project. The project team will meet weekly during the project.
A Project Board will be established to oversee RISE. This will be chaired by Gill Needham, Associate Director of Library Services, with membership including Judith Pickering – Project Manager for the OU DOULS project, Dr Tony Hirst – Lecturer in Telematics, and two Heads of Faculty Teams in the library, Judy Thomas and Clari Gosling.
Budget
This is a great project plan post, really clear and focused. I like the use of the pie chart for the budget.
Thanks Andy. I think the idea for the budget pie chart came from something you’d written about how to present the budget info. It worked out quite nicely.