The XCRI website has been running silent for a couple of months, not because of a lack of activity but rather the opposite. This post is an attempt to review and consolidate that activity, so followers of our work can appreciate where it has reached and where it is going.

XCRI's intention was to bridge the worlds of course marketing and course validation/quality assurance. Research into institutional practice and ambitions revealed a frustration that course information entered for one purpose was not readily available for use in another, and concern that the way a course was described for marketing purposes should match the way it was approved and the way it was delivered. XCRI has continued this investigation over recent months and has reached a conclusion that producing a course definition for validation/modification involves assembling fragments of information into a whole, whereas marketing a course involves communicating a serialized version of a subset of that assembly.

XCRI's R1.0 schema attempted to handle both course validation/modification and course marketing. As a consequence, it became somewhat unwieldy, and detailed analysis of its use across the UK revealed concentration on a limited set of the elements for the purpose of course advertising. Through discussions with early adopters, like Tavis Reddick of the Adam Smith College, Fife, we found interest in the notion of a Course Advertising Profile - a serialization of the larger, relational information model that was optimised for advertising and supported aggregation of learning opportunities by a national aggregator, such as UCAS, or specialist regional portals.

This notion of a specialised Profile, drawn from a larger information model, is very much in-keeping with emerging thinking on JISC's Reference Model work, under which XCRI is funded. XCRI has always pursued a development philosophy of publish early and seek commnunity feedback. It is this feedback on the R1.0 schema that has led to the development of a specialist Profile and an acknowledgement that more work is needed to construct a coherent Reference Model for the domain of Course/Curriculum Management, that encompasses advertising, validation and modification and supports identification of pathways for prospective learners and trails followed already. As Simon Grant's work has highlighted, there is a need to annotate course specifications with competence information so that learners can appreciate how engagement with a particular learning opportunity will enhance their personal development. There is also a need to articulate entry requirements in a way that facilitates lifelong learning and enables learners to appreciate further horizons of opportunity that open up on completion of a particular course of study. This area of the Reference Model is in its infancy. Institutions are at a much earlier stage of being able to articulate their processes and requirements for support in this area compared to course advertising. Identifying learner opportunities is computationally challenging. Prerequisite and corequisite rules governing curriculum assembly must be instantiated with learner enrolment and achievement history in order to determine whether a proposed study path is valid. Performing this calculation to determine all valid study paths from a particular point is a major challenge.

XCRI is committed to delivering something usable from each injection of funding it receives. Its initial 12 month funding produced the R1.0 schema, an XML repository demonstrator and some in-depth analysis and articulation of requirements, particularly the survey of 161 prospectus websites. With an additional 5 months of funding, XCRI has:

1. supported further R1.0 deployment and reviewed all R1.0 deployments
2. commissioned scoping work on competence mapping
3. commissioned a review of tools and technologies available to support authoring of curriculum content
4. begun to articulate a relational model underpinning assembly of curriculum definition documents
5. built on its review of online prospectus content and R1.0 deployments to produce a Course Advertising Profile - an XML specification developed from exisiting work that is optimised for course advertising
6. started to articulate a vocabulary of information fragments that are assembled to produce definitive documents and serialized to provide description for course advertising (these appear within the Course Advertising Profile schema)

These deliverables differ markedly from those envisaged at the start of the extension work. This is largely an acknowledgement that the domain of Course Information is unevenly complex and best tackled by picking off quick-wins, whilst simultaneously giving thought to the bigger picture.

Course Advertising Profile

The schema below is being released under a Creative Commons ShareAlike licence. Although the schema is annotated, full documentation is under development. The schema is released with two sample instances, drawn from the R1.0 deployments to demonstrate the Profile's range of applicability. The first is for a course provided by Adam Smith College; the second is for a short course provided by Oxford University's Computing Services.

Information Model for the Course Advertising Profile:

XCRI's infrastructure is being upgraded soon to include a code control repository, issue tracking and wiki documentation. Formal release of the Course Advertising Profile will include detailed guidance on deployment and a raft of examples demonstrating how the profile should be used with an aggregator. The information provided above is a sneak preview.

Relational Model for Assembling Course Specifications

Work is ongoing on the broader reference model for course information handling. Discussion documents regarding a fragment assembly model for authoring and maintaining course specification documents are shown below. Feedback is welcomed by the project manager (m.stubbs@mmu.ac.uk), and a more formal system of discussion forums will evolve when XCRI gets its infrastructure upgrade.

A Relational model that could underpin the assembly of course documents from fragments of relevant curriculum information

A More constrained model that attempts to articulate typical relations for entities in course documents: