For this post, we had to chose one “product” of the project, which we believed was to be most useful and reusable by others. We have done so many things over the last year that choosing one was almost impossible. After a lot of discussion and head scratching, we decided to promote as a product our collection of tools, examples and documentions, explaining the why and how of linked data, as well as the benefit one can get from deploying linked data in a higher education institution. We call this toolkit Tabloid: Toolkit ABout Linked Open Institutional Data.

Users

To clarify very quickly, the intended target audience for the Tabloid Toolkit are not the end-users of linked data. We focus here on helping people in higher education institutions with getting involved in promoting, implementing and deploying linked data within their institution. This includes more or less anybody who would have a role to play in the management of data and information, from PVCs to researchers, librarians and developers.

Overview

Tabloid is an evolving toolkit made of code, documentation and examples in various places, and trying to address the people with various roles involved in the deployment of linked data: from managers who want to quickly understand the benefits, to developers who are expected to work with it, develop applications and integrate it into their technical workflow.

In this sense, Tabloid can be seen an entry point to institutional linked data, with different parts being relevant to different people at different times. It includes many components distributed in different ways, and put together in a coherent structure in the Tabloid Page. In particular, the toolkit contains documentations giving an overview of the basic principles of linked data, of the way it concretely creates benefits and of simple examples of how such benefits can be exploited in research and education scenarios (see What is linked data?). It provides an overview of both the technical and organisational workflows that are necessary to deploy linked data in an institution, and provide some tool support to realise common tasks in such workflows. Finally, Tabloid puts a particular emphasis on the aspect of using and consuming linked data, providing documentation and experience reports regarding the use of linked data. It includes many pointers to a large variety of applications developed within the LUCERO project, together with reusable source code.

Link: The Tabloid page

LUCERO blog up to 1st July 2011:

Many parts of the Tabloid toolkit described above have been drawn out or described in blog posts on the LUCERO Blog. Here we give a brief overview of the content of the blog according to (mostly emerging) categories of posts:

Publishing Datasets

One of the major activities in LUCERO is related to the exposure as linked data of a number of datasets from the Open University. The posts in this category explain and describe how we realised such exposure for a number of datasets.

• PRONOM and Linked Data – This post is not actually about a dataset from LUCERO, but comments on one that was exposed by another organisation, based on the experience gained in LUCERO
• Publishing OpenLearn Metadata as Linked Data – OpenLearn is the repository of Open Educational Material from the Open University. This post explains how we expose the related metadata as linked data.
• Connecting the reading experience database to the Web of Data – This post describes how we exposed the content of the Reading Experience Database and built a demonstrator showing what can be achieved through connecting such data to external source.
• Publishing ORO as Linked Data – This post explains how the ORO repository of publications was exposed as linked data on data.open.ac.uk.
• First version of data.open.ac.uk – This post announced the release of the very first version of data.open.ac.uk with a small number of datasets.

Documentation and Support

The LUCERO blog is also used to provide easily accessible documentation regarding various aspects of the project. This category contains posts and pages that are intended to help people to better understand the principles and technologies related to linked data.

• What is linked data? – This page is intended to give a general overview of the principles and benefits of linked data for readers of the blog unfamiliar with it.
• data.open.ac.uk – This page gives an overview of data.open.ac.uk and of its relationship with the project.
• Example Queries – This post show example SPARQL queries helping users to understand what can be done with SPARQL and the data.
• Initial Overview of the LUCERO workflow – This post described the technical workflow applied in LUCERO to get from existing data sources to up-to-date linked data on data.open.ac.uk.

Tools and Applications

This category includes posts that describe tools and applications developed within the project. It is an important part of the activities in LUCERO, demonstrating through examples how one can benefit from linked data, and how to realise such applications.

• What to ask Linked Data? – This describes a tool that extract from a dataset the questions it is particularly good at answering. A demo shows the tool applied on 3 datasets from data.open.ac.uk.
• wayOU – Mobile Location Tracking App Using Linked Data – The application described in this post is a mobile app for OU users to track their location at the OU. It has won the “Best Demo Award” at the Extended Semantic Web Conference 2011.
• ROLE Widget Consumes Linked Data – This guest post from a member of the ROLE project explains how linked data available on data.open.ac.uk was used to create a widget for the learning environment created by ROLE.
• Results from the KMi Linked Data Application Competition – This post describes the 4 applications that were submitted to the linked data application development competition organised by LUCERO.
• Connecting the reading experience database to the Web of Data – This post describes how we exposed the content of the Reading Experience Database and built a demonstrator showing what can be achieved through connecting such data to external source.

Experience report – Guest posts

One great success of LUCERO is that it has managed to get people outside the project and the linked data community to engage with linked data, create applications of it and generally use the linked data we exposed for a variety of tasks. The posts in this category show a few of such examples.

• ROLE Widget Consumes Linked Data – This guest post from a member of the ROLE project explains how linked data available on data.open.ac.uk was used to create a widget for the learning environment created by ROLE.
• Know Thyself – This post written by a member of the communication services of the Open University shows how the availability of linked data can be used to quickly answer unexpected queries that aggregate resources from various resources.
• Putting Linked Data to Work: A Developer’s Perspective – This guest post written by a developer from the IT department of the Open University demonstrates how linked data can be used and integrated to write new and more cost effective applications, despite the initial confusion that linked data technologies often create.
• Introducing LUCERO – This post summarises the effort realised at the beginning of the project to explain and discuss with a large variety of people the expected benefits of linked data.

Project Plan

The 7 first posts on the blog gave the details of the project plan.

Hello World – This un-categorised post summarised, at the very beginning of the project, our expectations and plans for LUCERO.

Description of the Project

• Name: LUCERO – Linking University Content for Education and Research Online
• URL: http://lucero-project.info
• Dates: 1st June 2010 – 31st May 2011
• Programme: JiscEXPO
• Funding: £100,000
• Team: see About page
• Code License: LGPL
• Data License: CC-BY
• Content License: CC-BY
• Code and documentation: LUCERO on Google Code

The Lucero project Directly Incurred staff costs fund a full-time Research Assistant working in the Knowledge Media Institute. This post will be carrying out the key roles of extracting, converting and building the linked data datasets, creating the prototype search systems and carrying out the key technical work of the project. Overseeing this work will be the Project Director who has 20% of his overall time allocated to this project. Both have a high degree of experience in working with creating and exposing linked data.

The other Directly Incurred costs cover the work required to plan and extract library catalogue data from the Voyager library management system and to provide support to the project from metadata and Information Management experts. The Information Management expertise is expected to be of particular help in looking at embedding linked data in workflows and in information management practice.

The final element of Directly Incurred staff costs is represented by a 0.5 fte Project Manager. This post is essential in keeping the project on track and to budget.  Given the short nature of the project and the wide range of stakeholders it is essential that the project is managed effectively and that all reporting requirements are complied with.

The Directly-Incurred non-staff costs include a small amount to fund attendance at programme and dissemination events.

Directly-allocated costs include stakeholders allocated to Chair the Steering Group and provide strategic advice, liaison and guidance. An important element of the directly-allocated staff costs is the time from the managers of the key Arts Faculty databases that will be targeted as part of Workpackage 4. This time will be spent allowing the Project team to understand the data content within these databases, understand any licence or rights restrictions and consider the best methods of extracting and converting their data.

Finally there are directly allocated costs for Estates and indirect charges.

The project budget will be maintained by the Knowledge Media Institute Administration Manager Jane Whild in conjunction with the KMI Unit Accountant.

The Project Manager will be in regular contact with the KMI finance team to monitor and report on the budget to the Project Team and Steering Group.  An initial budget meeting has already taken place (30 June 2010) with the Project Director and Project Manager, KMI, Library and Arts Finance teams to plan and agree the budget management processes.

Timeline for WP1.

The second workpackage is concerned with the first task considered by LUCERO: to create the technical infrastructure necessary to expose University content as linked data and to semi-automatically create meaningful connections in this data, and with external resources. (WP2: Exposing University content as linked data)

Timeline for WP2.

Beyond the purely technological aspect of exposing University content as linked data, we will consider new procedures so that the practices of linked data are integrated within the current activities in a sustainable way. This is the goal of the third workpackage of the project. (WP3: Integrating linked data in University practices)

Timeline for WP3.

Exposing data as linked data is only the first step of the process, as it allows building a new kind of applications taking benefit from the meaningful connections established or derived from the data. The benefit brought by such applications constitutes the main incentive for the broad education and research community to integrate the principles of linked data. Therefore, the third workpackage of LUCERO concerns building demonstrators showing the interests of using linked data to both academics and students. (WP4:Demonstrating the value of linked data to researchers and students)

Timeline of WP4.

LUCERO’s benefit to end-users will be directly demonstrated through prototype applications (see Objective 3). However, the institution-wide benefits of linked data practices for data exposure and connection will only be fully understood in the long term, after the project is ended. Therefore, particular efforts should be spent as part of the project to ensure the sustainability and continuity of the activities started during the project, as well as to identify potential issues related to this sustainability. This also includes activities around the evaluation of the project and the dissemination of the project results, which are the topics of the workpackages 5, 6, and 7. (WP5:Sustainability in the exposure of linked data. WP6: Project Evaluation. WP7: Dissemination)

Timelines of WP5, WP6 and WP7.

Overall, the Lucero project will be managed according to the Open University Prince2 project methodology.  There will be weekly project team meetings with action points, regular Project Steering Group meetings, regular project reporting, and risk/issue logs to ensure the project achieves the aims and objectives set out for the project.

Dr. Mathieu d’Aquin (project director) is a researcher at KMi. He obtained a PhD from the University of Nancy, France, where he worked on real-life applications of semantic technologies to knowledge management and decision support in the medical domain. As a member of the EU Integrated Project NeOn, Mathieu has researched large-scale infrastructures for the discovery, indexing and exploitation of semantic data (e.g. the Watson Semantic Web search engine, Cupboard system for managing semantic information spaces), as well as in numerous research prototypes in concrete applications of these developments.

Fouad Zablith is a researcher and PhD candidate at the Knowledge Media Institute of the Open University. Within the LUCERO project, he’s working on modelling and deploying data in various university contexts, published within http://data.open.ac.uk. His research PhD is in the Semantic Web area, focussing on ontology evolution from external domain data, by reusing various sources of background knowledge. Fouad is also the web consultant of the Open Arts Archive project (http://www.openartsarchive.org), responsible of the implementation and maintenance of the website.

Salman Elahi is a research assistant at KMi. He obtained his Master’s degree in Knowledge Management and Engineering from the University of Edinburgh. Prior to joining KMi, he has been working as a Software Engineer on projects related to the use of semantic technologies to enhance search systems in the domain of Freshwater Sciences. At KMi, he is involved in the Watson and Cupboard projects. He has also started his part-time PhD looking at issues related to identity and personal information management.

Prof. Enrico Motta is Professor of Knowledge Technologies at KMi and a leading international scientist in the area of Semantic Technologies, with extensive experience of both fundamental and applied research. Over the years, he has authored more than 200 refereed publications and collaborated with a variety of organizations, including Nokia, Rolls-Royce, Fiat, Phillips, and the United Nations, to name just a few, while receiving close to £7M in external research funding.

Owen Stephens is the Project Manager for LUCERO. He joined the Open University in 2009 and was previously Project Manager for the JISC-funded TELSTAR (Technology enhanced learning supporting students to achieve academic rigour) project delivered at the Open University. Owen also works as an independent consultant to the library sector. As well as a strong technical background, he has been on the management team of the library services of two leading UK Universities he has been responsible for a number of innovative projects at both institutional and national levels. Owen was Project Director for the EThOSNet project to launch national e-theses service based at the British Library, and is the founder of the ‘Mashed Libraries’ events in the UK.

Stuart Brown (@stuartbrown) is Web Developments and Online Communities manager at The Open University. Involved in implementing the OU’s move to Drupal as default CMS Stuart is interested in both reuse of OU linked data within the OU’s web publishing environment as well as ensuring OU web publishing activity plays a role in the OU’s continued publication of linked data. Working with colleagues and systems across the OU Stuart hopes to help ensure that a linked data approach becomes a part of core OU activity.

Richard Nurse is Digital Libraries Programme Manager at the Open University Library.  Richard joined the Open University in 2009 and leads on Digital Library and website initiatives. He has considerable experience of library systems management and extensive experience of managing funded projects from the National Lottery and Wolfson Challenge Fund. Richard has been a key member of the recent collaborative JISC-funded TELSTAR (Technology enhanced learning supporting students to achieve academic rigour) project delivered at the Open University.

In addition to the core team, the project includes two sets of users and practitioners of linked educational and research data. This includes, on the one end, library specialists involved in managing data, especially about publications, course materials and archives (ORO, The OU Archive, the Library Catalogue). On the other end, out project plan includes the exposure of data from specific research projects, which include data modeling, requirement analysis and evaluation realized together with academics from 6 different projects at the Faculty of Arts of the Open University.

Deciding on which license to apply on the data itself is a slightly more complicated issue. One element that needs to be taken into account is the source of the original data. The ORO repository, library catalogue, archives materials, OpenLearn and iTunesU content is available for educational and non-commercial use, mostly under the creative commons licence. Only the publishable parts of the staff directory will be considered for linked data exposure. For new content and links, we are currently investigating the applicability of a variety of licenses such as CC0 and PDDL.

All content produced, including reports, blogs and documentation will be made available under the creative commons license with attribution.

1. through the benefit of developed linked-data based application to students and academics, i.e., how the deployed application have increased access to and usage of data in these communities.
2. through the successful application and deployment of the devised procedures within the Open University to create sustainable linked data exposure, i.e., having evidence that the rate and quality of linked data exposure will be sustained beyond the project, and that new applications will be developed (possibly by others) on top of this data.
3. through the adoption of the practices put in place within the project by other education and/or research organizations. Indeed, one of the major goal of the project is to document the experience acquired in setting up such practices within the OU in order to provide guidelines for others to engage with linked data.

Of course, the realization of such a project does not come without risks. Most obviously, like in many technological projects, the availability of key resources, technologies, skills and staff are essential. More specifically, linked data being a very recent set of technologies and principles, the risk of the necessarily underlying infrastructure not being ready and mature enough to support the ambitious goals of the project is non negligible, even if the project include some of the recognized experts in the area, with many connection in the research community. Finally, in relation with the relative novelty of linked data, important considerations will have to be given to the non-technical issues of exposing University resources, included the legal, business and ethical aspects that all raise their own set of specific issues for which very little experience exists until now.

Such risks and issues will be managed through the use of Risks and Issues logs maintained by the Project Manager, discussed at Project Team meetings and reported to Project Steering Group meetings.

Tremendous efforts are required for students for example to obtain an overview of relevant scholarly material concerning a particular topic of a course, or covering elements of some multimedia content. For example, answering a simple question such as “What courses are available that relate to this BBC programme I have just seen?” would require to manually locate the relevant resources, access them through many different systems and integrate the results. The same issues apply to researchers who rely on ad-hoc data collection, access and curation mechanisms, limiting their ability to flexibly exploit the
data, and to interpret it in connection with external information.

In addition, for researchers, the management and sharing of data is becoming a major issue. This is often realised through a database maintained by the project officer, manually entering and cleaning the
data, which is linked to a Web interface developed by an IT team. The main issues that arise from such a workflow include the inaccessibility of the data to other applications than its dedicated Web interface, the relative isolation of this data, being not connected to other information, but also that the relation between the database and its Web exposure is
non-trivial, creating a threat to its sustainability.

Through proposing clear procedures and technological support to the exposure of research and educational data as linked data, LUCERO will benefit “Users” and members of the Open University, especially:

• Course and programme teams, through more effective content collection for course and programme creation, as well as the ability to publish course content enriched with links to relevant (data) resources.
• Students, by providing multiple access points to educational and research data, as well as the availability of new tools to explore relevant resources.
• Researchers, through the availability of new data analysis tools for linked data, able to make emerge connections between previously unrelated elements on the basis of links to external datasets.
• The communication services, through new processes to realise the Web exposure of University content in an efficient and interlinked manner.

In addition, as LUCERO integrates the openness that characterize the Open University as well as, to a large extent, the linked data movement, within the exposure of educational and research data. Indeed, data published as part of the project will be made accessible freely and openly to any academics, student or research without restrictions.

Finally, besides the direct benefit of the open exposure of linked data, LUCERO intends to engage with the wider community by providing experience report, guidelines and reusable components, which can be picked up and employed directly by other organizations. As such, we intend to engage with the community of Semantic Web practitioners though interlinking and sharing of practices, as well as with the community of librarians/information managers regarding procedures to manage and expose linked educational and research data.

“What are the workflows, business processes, policies and technologies needed to expose the Open University and related digital content as linked data?”

“How can we integrate linked data technology in a sustainable way to support the research and educational activities of a Further or Higher Education organisation?”

LUCERO collaborates closely with the Open University Faculty of Arts to prototype and evaluate specific content exposure and linked data applications for researchers working within the Arts and Arts History domains, providing experience on the exposure and connection of research data outputs, and demonstrating their concrete benefits.

Exposing resources as linked data creates a potential for broader reuse of their content, impacting on potentially large numbers of students and research communities. In LUCERO, we aim to document business process changes required to achieve successful integrated institutional approaches and behaviours required to facilitate content and data reuse alongside documenting the development of policy and recommended standard-based, semantic technology interoperability solutions to support the effective delivery of educational and research data linked exposure.

More precisely, the planned outputs of the project are:

• The deployment, test and documentation of a technical infrastructure, a toolkit, to facilitate the creation, exposure and use of linked data, implemented within the Open University, but designed to be reusable in other HE/FE institutions. This includes the realisation of interfaces to data creation, storage, publication and semi-automatic linking suitable to be used by library staff and academics, and that integrate with their current working environment.
• The identification, documentation and validation of the processes necessary to integrate linked data in the Univeristy’s practices and workflows, including in particular the business, legal, ethical and organisational aspects.
• Demonstrators of the benefits of exposing educational and research data as linked data through the realisation of applications improving access to educational and research data in the domain of Arts, for both researchers and students.