13th CETIS Enterprise SIG Meeting: XCRI Showcase
The 13th SIG Meeting was held in the new Isis Room at
Oxford University Computing Services
(OUCS), hosted by Adam Marshall and staff at OUCS.
Intro: About the XCRI Project
The meeting showcased the work of the
XCRI (eXchanging Course
Information) Project, which arose due to a need identified by members
of the Enterprise SIG (which focuses on transferring learner data
between college systems and between institutions), to develop a
standardised way of transferring prospectus or course information.
In January 2005, a group of people from the SIG, plus members of UCAS (the
UK Universities and Colleges Admissions Service, which processes applications
to study undergraduate courses), and a representative from JISC, met in
Manchester to discuss how to proceed with tackling this issue. As chance would
have it, JISC were looking for bids for their Reference Models call, and
encouraged the group to submit a bid by the call closing date the following week!
Under the leadership of Project Manager Mark Stubbs (from Manchester Metropolitan
University), a bid was submitted and accepted, and the XCRI Project got underway.
Julie Hardman (XCRI Research Assistant) surveyed 161 institutions throughout the
UK, to identify common course elements needed by the proposed XCRI schema,
and Mark and other interested parties met together to begin developing the schema.
Mark also developed a highly commended project website, featuring dozens
of document links and a blog outlining the project's progress.
The earliest version of the schema was released just before Christmas 2005,
and by early 2006, several different individuals had already began to use
the schema. In March 2006, the first funded phase of the project came to an
end, and it was decided to hold an XCRI showcase meeting to enable Mark
to report back on the project, and hear the experiences of the early
implementors of the schema.
Introduction & Introductions (Powerpoint)
Vashti Zarach, CETIS Enterprise SIG Coordinator, opened the meeting, welcoming
new faces and Enterprise SIG regulars. She apologised for the packed agenda,
which was a reflection of the large amount of work done with XCRI, and explained
the aims of the day: namely to present and disseminate the work of the project.
Meeting attendees then gave their names and reasons for attending: many of the
delegates were attending to give presentations on their involvement with the
project, and the others wanted to learn more about XCRI. Bill Olivier from
JISC was attending to assess the effectiveness of Reference Models as a way
of enabling community development.
One delegate's reason for attending: "I've been waiting a long time for
something like this to come along."
Mark Stubbs: The XCRI Project
Mark explained the purpose of XCRI: to provide a standardized format for
course data used for marketing courses to prospective students, to ensure
that the courses delivered by institutions matched the courses advertised to
students, to populate MLEs with course information for students enrolled on
courses, and to supply data for funding and course awards.
Mark explained the roots of the project in discussions at Enterprise SIG
Meetings, and the progress of the project via "summits" (meals out in
restaurants after meetings!) and huge amounts of community collaboration.
Finally, he outlined the project outputs: a very active website and blog,
the first version of the XCRI schema, and a number of XCRI deployments by
different people. Mark explained that the project team had looked closely
at the Norwegian course schema, CDM, but had decided in the end to develop
their own schema as there were some issues which they wanted to tackle
differently. Mark concluded by explaining that the rest of the day would
focus on various deployments of XCRI. The deployments map on his Powerpoint
shows just how far across England and Scotland XCRI has spread in just two and
a half months.
Quote from delegate on Mark's decision to put a project blog on the
XCRI website: "A very good decision!"
ioMorph & XCRI (Powerpoint)
Selwyn, Phosphorix developer, gave a presentation on the newly released
ioMorph, which he described as a "transformation service". In essence,
this means that you can input course information data into ioMorph, and
it will return the data in XCRI XML format. For example, data can be sent
from a Student Record System through a plug-in into the ioMorph web service,
and be returned as XCRI XML.
Liverpool Hope University has developed an ioMorph plugin as part of the
Learning Matrix project. Selwyn demonstrated how he could select a course
from the list at Liverpool Hope, send the data to ioMorph as a long string,
and have XCRI format data come back.
ioMorph can handle both student and course information.
the ioMorph toolkit here and
Ben Ryan: Curriculum
Documents - Creation & Storage (Powerpoint)
Ben Ryan began work as one of the team on the
Progression (Mapping Curricula to Facilitate Student Progression)
project around the same as time as Mark began XCRI, and Balbir Barn
COVARM (Course Validation Reference Model) project. Recognising they
had common ground, the three projects made contact and engaged in some
collaborative work. Ben has been involved with XCRI from the early stages
of the project.
Ben talked about curriculum documents: the process of creating records
describing courses. He explained that the process of developing curriculum
documents had begun with an analysis of a wide variety of documents about
courses created by colleges (web pages, word docs, etc). From this, they
had moved on to working on two versions of curriculum documents, one web
based and one word style document. He demonstrated the web based documents, which
displayed details of intended course outcomes and course assessments. It
is possible to use this information to create one page reports showing
learners' assessment results and completion of course outcomes.
Ben then demonstrated how he has enabled XCRI course information in XML
format to be converted into a word document using Word ML. I don't know
exactly how this works, so you'd have to contact Ben for more info, but
the conversion worked at a click, changing a string of XML into a word
document with paragraphs of info about a course. The audience was
audibly impressed at the conversion! Ben explained that this process
could be easily used to create customized course documents,
presenting different information to different people.
Quote from SIG developer: "I dream of having freely available open
source software for creating authority tools for courses."
Anthony Beal: The
Learning Matrix Project (Powerpoint)
Anthony discussed the Learning Matrix Project, which offers taster packages
of Higher Education courses to encourage non-traditional HE learners to
enter Higher Education. The courses begin with a face to face session in
local colleges or universities, then learners spend some time online,
and some time back at the college again, as many of them can initially
feel nervous about the electronic component of the course and value the
opportunity to meet tutors and ask questions face to face.
The presentation then took us through the process. Firstly, course information
details are added to the Learning Matrix by an institution. Learners log
into the Learning Matrix portal and view courses. PDP services built into
the portal encourage students to explore their learning needs and
existing skills. They apply for the course from the portal, and, if accepted,
are enrolled at the institution. Once the course is finished, the tutor
confirms course completion, they are sent a completion certificate, and details
of the completed course are added to their Learner Record.
The Learning Matrix project has been using XCRI to format course information
for the portal, so that courses from different institutions can be described
in a standardised way. Anthony made the important point that we need to
consider who XCRI course descriptions are for, e.g. course information
for academic purposes can be quite dry, but to market courses to students,
the information needs to be presented in a more interesting and enticing way.
Finally, Anthony raised a few issues for consideration: namely the need to be
consistent about XCRI field entries, the need for an authorised vocabulary
to use when filling course information fields, and the need to consider a
way to describe items which aren't simply courses, such as Open Days. Mark
was very grateful for the feedback, and explained that Specific
Vocabularies was one of the workpackages proposed in the current bid for
further funding for XCRI.
Afternoon presentations: Case studies from people using XCRI
Alys Morgan & Keith Lewis, Oxford University:
Alys Morgan: Skills Training
Information (Powerpoint) /
Keith Lewis: XCRI at
Alys Morgan gave the first half of the presentation, explaining that
her job involves producing a central information database about skills
training courses at Oxford, which typically tend to be short courses and
workshops. Using a diagram, she demonstrated how the information she
received to compile the data about these courses came from a wide variety
of different sources, in a variety of formats. Oxford University
Computing Services (OUCS) had decided to start formatting and sending their
information in XCRI format, and it was hoped that XCRI could potentially
be used to unify the existing system by encouraging other sources to send
data in this format.
We then heard from Keith Lewis at OUCS, who had been using XCRI to encode
course information about short IT skills courses, in order to transfer
information from their courses database. Keith very helpfully gave the
XCRI Project team a list of implementation issues which he had encountered.
As the project has only been running for a year, the team were very pleased
to be getting critical feedback so quickly!
Keith singled out the lack of documentation as a particular problem for early
implementors, and Mark and Ben explained that there had been some
documentation with the draft version of XCRI, and that lack of documentation
was partly due to the schema having only just been released. Mark
made a note of the issue for future work. The lack of a heading for "cost"
or "charge" in the schema was also mentioned, as was the need for an XML
editor which could distinguish between required and optional elements
when filling in course details in the schema.
Keith concluded that the schema does succeed in its mission, but is complex
to use, and needs more model examples, documentation and development. He also
encouraged the team to promote XCRI in e-Government circles and try to
get DFES to take up the schema.
Quote from team member Ben: "All of your comments are valid - it's
been adopted stupidly quickly!"
Quote from Mark: "It's superb feedback and you've done a remarkable
job considering the lack of documents!"
Michael Aherne: Producing XCRI for an FE College (Powerpoint to be
Michael Aherne was the first person to attend a SIG meeting and raise the issue
of prospectus information standards, back in
SIG meeting 6 in
North Wales. He has been involved with the XCRI project from the start, and
as soon as the first schema was released, set out to test and prove his
goal of having a course information standard which could be used by an FE
college with limited personnel and resources.
Michael's experience bore out his theory. Reid Kerr, like many FE colleges,
has just one web developer - Michael, who mapped their existing Access
course database to XCRI. He found the mapping process pretty straightforward,
as the existing database mapped easily to XCRI, and it was fairly obvious
which information to put in the XCRI fields. Michael demonstrated some
online course information in html format, which was then changed into
xml and gathered together in XCRI tags.
Issues identified by Michael included a lack of vocabularies, and problems
with institutions using different names for the same course, to which
Mark replied that vocabularies was one of the workpackages mentioned
in the application for funding for the second phase of XCRI. Michael also
highlighted the need to involve the Scottish Qualifications Authority
for XCRI to be adopted successfully in Scotland. Overall, however, he
believed that XCRI could be implemented quickly and easily by a single
web developer in an FE college.
Quote from Michael: "The whole process (apart from reading
and understanding the schema) took about two hours."
Using XCRI to transfer entry requirements to UCAS (Powerpoint)
Alan Paull, currently a consultant, has worked with course information
since 1979, and was involved in the XCRI Project from the beginning as
a Programme Manager at UCAS. UCAS
is a UK organization which processes undergraduate applications to
universities and colleges, and therefore deals with an enormous amount
of course information from across the UK.
Alan's presentation focused on his work mapping course information on
entry requirements at Plymouth University to XCRI, and then mapping XCRI
to the UCAS courses database, to enable information to be transferred
from Plymouth to UCAS. Mapping the information between Plymouth and XCRI
was relatively straightforward, but mapping data from XCRI to UCAS was
more complex, and took about 3 days. Alan again stressed the importance
of specific vocabularies for XCRI in speeding up this process.
The mapping was undertaken using the tools XML Spy and MapForce
(for visual mapping), and Alan showed us a screenshot of the mapping
process in action with MapForce.
Overall, Alan concluded that the work proved it was possible to map from
the Plymouth course information to XCRI, and from XCRI to UCAS, showing
that the schema seems to be flexible enough to cope with a variety of
HE course information formats. However, he stressed that the schema was
currently complex to use, and more focus on documentation and training
Quote from Alan's slides: Suffered from the "early pioneer syndrome".
XCRI Query Web Service Demonstrator (PDF)
Paul Walk is a longstanding Enterprise SIG member and web developer who
has also been involved in the discussions about standardising course
information from the very beginning. Paul decided to create an XCRI
demonstrator, demonstrating XCRI query services, which other developers
could use and explore. Just a couple of hours after releasing the
demonstrator, Scott Wilson had consumed services from the demonstator
using the Personal Learning Environment (PLE) Reference Model demonstrator: i.e
gathered course information from Paul's demonstrator in response to queries
from the PLE demonstrator and displayed them in the PLE.
More info on the demonstrator, and consumption of services by PLE on
the XCRI blog.
Paul gave a live demonstration of the XCRI demonstrator carrying out
queries such as "return information about course outcomes from all
modules", and "return details about all modules".
The XCRI Query Web Service
is online here.
At the end of the day's presentations, Mark took the stand again, and
concluded that the XCRI project had come a long way in 12 months, but
that there was still a long way to go to achieve sustainability. He
proposed that the next stage of development needed to move up a gear,
and that this could hopefully be done if the bid for extension work
on XCRI was successful. He outlined his commitment to tackling some
of the issues raised during the day, such as defining clearer vocabularies,
and thanked everyone for the community participation which had enabled the
project to achieve so much so quickly.