Enterprise Meeting 17

Back to Previous Enterprise SIG Meetings: 2008.

Introduction
The 17th JISC CETIS Enterprise SIG Meeting was held on Friday 4th April 2008 at the University of Bolton, and focussed on scheduling and integration of personalised systems for students.

Back to Top of Page

The COVa (Course Validation Arranger) Project (PDF Format - 589Kb)
Balbir Barn, Thames Valley University.

The The COVa (Course Validation Arranger) Project is exploring the use of BPM (Business Process Management) toolsets for defining and managing administration processes. It came out of the COVARM Project, which aimed to produce a reference model for a "Course Validation" service for the JISC (Joint Information Systems Committee) eLearning Framework Programme.

One result of the COVARM Project was the discovery that the business processes involved were very complex and too large to easily handle. So it was decided to deconstruct the process into more manageable areas, using the notion of "events". COVARM had initially tried to implement the whole process but it turned out to be far too large. COVa on the other hand concentrates solely on two areas:


 * Exploring the technological issues by preparing a programme proposal;
 * Setting up a validation event.

The setting up of a validation event has used various document sets, such as the UML (Unified Modelling Language) Model from COVARM. A BPEL (Business Process Execution Language) process was defined for the sub-process and will invoke the operations required to implement the process.

However, horrendous issues have been discovered with this approach. Going from BPEL and WSDL (Web Service Definition Language) was problematic as it didn't seem possible to move the structures across (although this may have been a problem with the Oracle server). Also, as multiple tools were being used, a lot of round-trip engineering was being done that actually wasn't very round-trip! So, it seemed that the toolsets being used did not allow for good implementation. XDE was used for the model transformation and there were lots of multiple transforms. Plus, it wasn't feasible to hard the application using BPEL as it would have required an army of programmers.

The BPMN (Business Process Modeling Notation) 1.0 Specification allows emerging toolsets to put everything together from analysis to implementation. It really seems to have taken off as a specification and has around 300 pages of notation. The BPMN consists of flowchart notation for defining business processes and can be mapped to BPEL. The notation, which is semantically complete, is split into events, flows, pools, swim lanes, etc and should cover all types of complexity (see the COVa Website for more information). However, there is no guidance or methodology on how to use BPMN and most of the toolsets haven't implemented all of the notation, because it is so large. BPM is important for understanding functions but it can be taken further. To date, the eFramework has neglected process management and the COVa Project Team are trying to influence its use within the eFramework.

The COVa team did a literature review in order to discover how a toolset might be chosen. The toolset had to fit certain parameters - it had to be open source, cheap, and supported. It was decided to use the Intalio toolset, although support for the free version has been patchy. As with other toolsets, the full BPMN is not available, so it's not easy to draw collaborative activities in Intalio and the sharing of information between different roles is also difficult. Although the BPMN represents delay/timer events, Intalio doesn't, so a workaround had to be invented. Also, semantics are currently not available in the toolset.

The process involved taking existing COVARM models, and re-importing them into the Intalio toolset (translating them into BPMN). They were then implemented using COVARM services (consume COVARM WSDL, form design in Intalio, map complex XML (eXtenisble Mark-up Language) data from forms to XML struture of services) and evaluation and executed in the toolset in order to avoid problems moving between toolsets.

One issue with the BPMN itself is that is won't allow data to be shared across pools (i.e. across organisations or institutions). This has been very constraining and has caused problems. However, a workaround was developed by using a particular naming convention of roles (i.e. "Role: Subrole" which could translate to "University: Registry" or "University: Computing Department", for example), which fools the BPMN into seeing the different organisations as one pool. In this way it was possible to get messages to flow across the swim lanes.

It was hoped that a BPMN toolset would make life easier. To some extent it does as it provides a good way of capturing the needs of a system, particularly if a subset is chosen, but it is still quite hard. Toolsets, like Intalio, allow you to look at business needs and then take them forward to implementation, however, it is still a big leap. It might be worth trying proper commercial toolsets rather than commercial ones.

As part of the project, the COVa Team have been running BPMN Workshops and have produced guidance notes on process modelling (currently available as a draft with an extended abstract).

Back to Top of Page

Building a Network of Course Information with CourseExchange and ioNetworkNode
Selwyn Lloyd, Phosphorix.

Phosphorix will publicly release CourseExchange and ioNetworkNode at this year's JISC conference. This presentation consist of practical insights in how to use the software by demonstrating pre-recorded use of the CourseExchange agent, hub and portal... We were also  demonstrated on how to install and configure ioNetworkNode.SIG Members questioned and explore the problems which can be solved by ioNetworks for course information at the end.

The session on Phosphorix focuses on open source statement with respect to two lively demos on ionetworknode and course exchanges.The phosphorix website had changed over the last 12 months.Many projects are based on the ionetworknode.The  company found Wikipedia a better place to put the ionetworknode.

Open source statement
The main aim of it is any organisation can take software and develop it and is completely free.They have figured out the best policy for the open source statement by an APACHE2 license for ioMorph,ioNetworknode.The transformation code was not open sourced.The main aim of the ionetworknode was to setting up the relations between institutions.More details on ioMorph and  ionetworknode

CourseExchange
The course exchange comprises of three parts namely the course agent,course hub and the ioportals.The ionetworknode may host one or more course exchange web application with one hub per network.The ioportals have the ability to connect to one or more ionetworknodes and coursehubs.For more info CourseExchange

Back to Top of Page

The PADDLE (Piloting A DecentraliseD Learning Environment) Project(Powerpoint Presentation)
Henry Blackman, University of Chester.

The PADDLE (Piloting A DecentraliseD Learning Environment) Project is exploring the uses of personal learning environments/user owned technology, and how this can pull technology delivery from IT departments into the hands of the user.

The University of Chester has its own in-house VLE (Virtual Learning Environment) - IBIS - comparable with a commercial VLE, and an in-house portfolio system. IBIS uses TinyMCE for in-house editing. The team wanted to explore what a PLE (Personalised Learning Environment) means to the institution - not just the technology but about what users actually see. For example, are students using personally chosen sites, such as blogs, MySpace, etc? Is a PLE just about changing the delivery mechanism? If so, it could be a good thing anyway as it allows the students to access what they want, where and when.

The students trialling the PADDLE Project are non-traditional, work-based, mature students on a Health Informatics Foundation degree and although they are IT (Information Technology) literate, they are not IT students. Therefore, the project focus is on content delivery and as the students are busy in their jobs, it is important that all information is "push" driven and is web, mobile, or desktop based. The course is funded by the NHS (National Health Service), so certain information has to be delivered at certain times. The course modules have to be done in a particular order and are time released, building on the previous week's work. This is being implemented in the form of "home pages", which are being developed using existing applications as there is plenty of choice. The PADDLE Project allows content to be created, sequenced, time released and then sent as an RSS (Really Simple Syndication) feed to the student's home page. Security is covered by a token and user name approach to check authenticity, although IBIS is only available to institution members anyway.

The personal homepage uses NetVibes and has tabs for home, work, and study. The NetVibes UWA (Universal Widget API) allows widgets to be created for all sorts of platform, e.g. 'phone, desktop, etc. People are used to working with widgets, e.g. iPhones, Vista sidebars, etc, so you might as well use what's there rather than recreate something. NetVibes is an all-in-one development framework and is very DOM (Document Object Model) oriented.

There have been some issues. For example, the team decided to use RSS 2.0 but many readers don't yet support some of the objects that need to be displayed. Forms don't always appear when included inline and it's difficult to embed items and an alternative means of disply has to be found. One way round this is to include a box saying "Click here" which goes to complex content, such as video, in the VLE. Although it's clunky, it does work. Also Safari tends to put new feeds into a long linear list, which are then all delivered at the same time. There is limited mainstream support for iCAL and FOAF (Friend of a Friend) (although the decision as to whether to use this technology or not still has to be made).

As the students are not IT students, they may only have limited IT experience, so an induction and pre-course training is provided. However, it can be hard to sell the idea of a personal homepage to the students. Why should they use it rather than the VLE? Because of the limitations of the RSS feed, students often find themselves back in the VLE, so why shouldn't they stay there in the first place?

The team want to make sure that the user experience is the best that can be provided. Up until now, 6 or 7 URLs (Uniform Resource Locator) have been given for each course and a student could be on 6 or 7 courses. Instead, there now a move towards widgets, so that IBIS becomes a portal and students can use whichever widgets they prefer. The focus on microformats allows the students to choose the technology they want for the delivery of particular types of information, from calendaring to e-mail.

Academic staff generally use the VLE to create content and it is expected that this will continue to be the case, even with the move to widgets, for accessibility and other reasons. The project is currently into the first year of its two year run and it is hoped that it will continue.

Back to Top of Page

Timetabling and Resource Scheduling in the University of Bolton(PowerPoint Presentation)
Andy Dale, University of Bolton.

Presentation on the consideration in curriculum development of room and resource needs, such as the need for particular sorts of learning spaces and equipment; shared calendars enabling staff and students to check the availability of, and book, rooms and resources needed for learning activities in the University of Bolton.

Andy Dale gave a presentation about use of CELCAT for timetabling in Bolton. The software has a number of modules including a SAT (Student Attendance Tracker) which helps with student retention. There is also a Notifier module, which is used to notify people of any room changes and an AutoCal calendaring system.

The basic timetable is created in CELCAT, with staff details imported in from HR (Human Resources). Information on course modules is updated daily. It is then imported into SITS and then into CELCAT. Students can select modules for the next academic year in March or April and it is hoped that students will be able to produce individual timetables for themselves from September 2008.

The Roombooker module, which will also be rolled out in September 2008, will allow participants to be invited to meetings and identifies what facilities are available in each room. The information can be e-mailed or sent via SMS (Short Message Service).

Back to Top of Page

Discussion
Led by Sarah Davies, JISC.

Towards the end we had a small discussion around timetabling and resource scheduling. JISC have also just put out an ITT (Invitation to Tender) "to conduct a study on the administrative processes and challenges around timetabling and resource scheduling for learning opportunities, and the way in which these link with curriculum and session planning". Finally, We have requested the SIG members to input their views to theenterprise mailinglists.

Back to Top of Page