Pedagogy Forum
Contents
Events
Forum Details
About the forum
Contact or Join the Forum
Resources
CETIS Learning Design Workshop Report
CETIS Learning Design Workshop 28th and 29th October 2003
Report
By Sarah Currier, Lisa Corley and Rowin Cross
Facilitators: Lisa Corley (CETIS Pedagogy Forum Coordinator)
and Sarah Currier (CETIS EC SIG Coordinator)
Presenters: Oleg Liber & Bill Olivier (CETIS Directors),
James Dalziel (WebMCQ/Macquarie University/IMS Australia), Jocelyn Manderveld
(OUNL)
Note-taking and other support: Rowin Cross (CETIS Assessment
SIG Coordinator)
Hosts: Gerry Graham and Sheila MacNeill (Learning and Teaching
Scotland)
Participants: Murray Allan (SFEU); Sandy Anderson (Napier University); Neil Ballantyne (University of Strathclyde); Phil Barker (Heriot-Watt University); Phil Beauvoir (Bolton Institute); Frank Benneker (Universiteit van Amsterdam / Surf); Luke Bennett (Becta); Jennifer Booth (National University of Ireland, Galway); Victoria Boyd (University of Durham); Sandy Britain (JISC); Joanna Brown (Leeds University); Shelagh Brumpton (Newark & Sherwood College); Colin Calder (University of Aberdeen); Gayle Calverley (University of Manchester); Clive Church (Newark & Sherwood College); Lorna Campbell (University of Strathclyde); Grainne Conole (University of Southampton); Edd Dawson (University of Derby); Peter Douglas (Intrallect); Cuna Ekmekcioglu (University of Edinburgh); Rachel Ellaway (University of Edinburgh); Jim Everett (Stevenson College); Steve Fox (University of Edinburgh); Andrew Gold (University of Salford); Pierre Gorissen (Fontys University of Professional Education, The Netherlands); Paul Hollins (Bolton Institute); Dan Howell (Kilmarnock College); Allison Littlejohn (University of Strathclyde); Marion Manton (University of Oxford); Colin Milligan (University of Strathclyde); Trish Murray (University of Durham); Howard Noble (University of Oxford); Daniel Owen-McGee (University of Derby); Sharon Perry (Bangor University); Mark Pettigrew (Sheffield Hallam University); Mark Power (Bolton Institute); Peter Radcliff (University of Derby); Pete Rainger (TechDis); Judith Ramsay (Glasgow University); Paul Riddy (Southampton University); Karen Saunders (NEC); Bernard Scott (Cranfield University); Paul Sharples (Bolton Institute); Nadia Spang Bovey (University of Lausanne, Switzerland); Tony Toole (Coleg Sir Gar); Stephen Wan (Sheffield Hallam University); Lipeng Wen (University of Hull); Sue White (Coleg Sir Gar).
Apologies: Helen Beetham (JISC); Catherine Durkin (University of Strathclyde); Carol Higgison (University of Bradford).
Introduction
Throughout 2003, both the CETIS
Educational Content SIG and the CETIS
Pedagogy Forum experienced a surge of interest in the new IMS specification
IMS
Learning Design v1.0 (IMS LD), which was publicly released that February.
Presentations and updates on the rationale behind the specification and its
development had been piquing interest prior to this, but when the IMS LD-inspired
tool LAMS
(Learning Activity Management System) was demonstrated by James Dalziel
at a number of meetings and workshops throughout the UK, it struck a chord with
academics and learning technologists alike, giving them a practical taster of
what a full implementation of IMS LD might be able to achieve. There was increasingly
a feeling that the promise of IMS LD to enable more flexible, pedagogically
diverse delivery of e-learning may actually bear fruit.
Meanwhile, the OUNL (Open University of the Netherlands), the originators of IMS LD’s precursor EML (Educational Modelling Language), announced that development of EML would cease and their expertise would be brought to bear on the ongoing development and implementation of IMS LD. They launched an online community, Learning Networks, to support this and related e-learning issues. Resources include a Learning Design FAQ; space for the IMS LD developers’ network the Valkenburg Group; and an online discussion facility. The EML player Edubox, developed at the OUNL, was taken into production there in September 2003. Edubox is at this moment being used for nine courses with in total 2200 students. The speed of scaling-up and further development of Edubox is now subject to a decision.
In the UK, CETIS’ funding body the JISC had funded the RELOAD project as part of its X4L (Exchange for Learning) Programme. Part of RELOAD’s remit is to develop an editor and a runtime delivery system for IMS LD. RELOAD was therefore interested in learning about what its user community’s needs would be from such a tool.
So, in order to meet both the insatiable desire of the e-learning community for more on IMS LD, and the need of the IMS LD development and implementation community for end user feedback, the EC-SIG and the Pedagogy Forum organised a two-day workshop on IMS LD. This took place at Learning and Teaching Scotland in Glasgow, on 28/29 October 2003. It was free, and open to all interested parties involved in UK education. It was attended by 57 participants, mainly from UK HE & FE, but also including one participant from Switzerland, one from Eire, three from the Netherlands, and presenter James Dalziel from Australia.
The workshop also saw the announcement of a new CETIS Briefing Paper on IMS Learning Design, being written by Ann Jeffery of Huddersfield University and Sarah Currier, the CETIS EC SIG Coordinator. Participants were given a draft copy to read and comment on; feedback was incorporated into the final version, which was launched at the 9th EC SIG Meeting in Manchester. It is now available for free download as a colour PDF document from the CETIS website. This document is aimed at beginners – it is worth reading first if this workshop report is your first foray into the world of IMS LD.
The Agenda - Lisa Corley
(Presentation
slides)
Lisa welcomed everyone to the workshop and noted the importance of IMS Learning
Design:
As IMS notes:
"The development of a framework that supports pedagogical diversity
and innovation, while promoting the exchange and interoperability of e-learning
materials, is one of the key challenges in the e-learning industry today".
(From the Learning Design Best Practice Guide: http://www.imsglobal.org/learningdesign/ldv1p0pd/imsld_bestv1p0pd.html
)
Lisa gave a brief overview of the agenda for the two days and after a welcome to Learning & Teaching Scotland from Gerry Graham, the two directors of CETIS, Oleg Liber and Bill Olivier, opened the workshop with presentations introducing IMS LD.
Why Learning Design? - Oleg Liber
(Presentation
slides)
Oleg began by looking at the pedagogical rationale behind IMS LD, i.e. what problems did it set out to solve? He gave an overview of the IMS approach to interoperability specifications from 1997 on. IMS began by developing single-issue specifications such as IMS Content Packaging, with the support of the commercial VLE companies with IMS membership, whose products were aimed at the mass market, and who emphasised a traditional pedagogy of “read, discuss, test”. Meanwhile, interest in interoperability specifications was growing in the UK; where there was an increasing desire to develop tools and specs which would support a more diverse range of pedagogies. A UK IMS Centre was mooted; this Centre eventually became CETIS, which represents UK HE/FE on IMS and other standards bodies. Oleg Liber and Sandy Britain were commissioned by the JISC to report on pedagogy and VLEs, and in 1999 produced the seminal JTAP report “A Framework for the Pedagogical Evaluation of Virtual Learning Environments”, which looked at evaluating VLEs in the light of Laurillard’s conversational model of education, and Stafford Beer’s ‘viable systems model’. Finally, in February 2001, IMS formed a new group looking at interoperability for learning design. This group went in search of a model, and the OUNL’s Educational Modelling Language became the basis for the development of IMS Learning Design v1.0, with the assistance of Bill Olivier and the CETIS community.
From EML to LD - Bill Olivier
(Presentation
slides)
Bill is co-author of the IMS LD specification as well as co-director of CETIS. He outlined how the problems of describing and sharing complex pedagogies were tackled- i.e., how the OUNL developed the Educational Modelling Language, and how this was taken up by IMS to create IMS LD. He finished by explaining how IMS LD works.
The PowerPoint slides include extensive notes and present:
- Some background on EML at the OUNL and its transformation in IMS into the Learning Design specification
- How Learning Design fits into and extends Content Packaging
- The main parts of Learning Design
- The dynamics of Learning Design
- A slide on the Versailles Experience role-play example in the BPIG
(the BPIG shows how this is translated into LD XML) - A quick look at Learning Design Level B & Level C
- How Content Packages and SCOs (Shareable Content Objects) are included and Simple Sequencing is accommodated in a learning design
- An extended example of how the dynamics of a project-based learning scenario would play out in LD
- It ends with a look at the support activities of the Valkenburg group and UNFOLD, an EC Framework 6 funded Support Action geared to facilitating the implementation, interoperability, effective use and adoption of Learning Design
Creating & Delivering Courses Using IMS LD: Two Practical Sessions
After lunch the workshop split into two groups of ca. 25 people each for two-hour practical sessions. These included a hands-on session with the LAMS tool in LTScotland’s computer lab, and a session with Jocelyn Manderveld of the OUNL, going through the process of developing and delivering a learning design using the EML player Edubox. The following day, the morning was taken up with the same two two-hour sessions running simultaneously; workshop participants were therefore able to do both sessions. They had the opportunity to do the more theoretical OUNL one first, followed by the practical LAMS session, or vice versa. The workshop organisers were grateful to find that the learner preferences of participants split the numbers almost exactly down the middle!
Hands-on LAMS (Learning Activity Management System) demo- James Dalziel, Macquarie University E-Learning Centre of Excellence (MELCOE) / WebMCQ.
LAMS is a tool inspired by IMS LD, although it is not a reference implementation of the specification. It had been demonstrated at a number of recent CETIS meetings, sparking intense interest. The developers of LAMS contribute development issues to the IMS LD community known as the Valkenburg Group, and the results of this workshop will be part of that process. This exercise was the first ever group hands-on LAMS session anywhere in the world.
James began with a brief presentation about the rationale behind LAMS and a demo of the tool.
There is also an article by James Dalziel about LAMS at http://www.melcoe.mq.edu.au/documents/ASCILITE2003%20Dalziel%20Final.pdf
James explained that the problem of e-learning development being driven by technology instead of pedagogy (rather than the other way round) was a key reason for the development of both IMS LD and LAMS.
While one way of understanding IMS LD is via the analogy of a play, with acts and actors, the approach taken in designing LAMS was that of looking at the three components of a learning activity:
- Who is involved?
- What content is needed?
- What tools are needed?
Constructivism activity
James showed a simple example- an activity based on the question “What is constructivism?” One way of presenting this to learners is:
Week 1: A discussion environment – the class share their initial thoughts.
Week 2: Review of content provided by the teacher, and search for websites via Google, with a “framing” of the Google search page with instructions by the teacher.
Week 3: Small group debates with scribe reporting back to whole class.
Week 4: Submit report for marking and comments.
James demonstrated the students’ view of this sequence in LAMS, followed by the monitoring, or tutor’s view, and finally showed how this activity was created in the LAMS authoring tool, and how it could be changed. Workshop participants then took part in this exercise as learners in small groups. Finally, they were set loose to create their own learning activities on LAMS to see how it works as an authoring environment. As this just whetted appetites, James allowed participants guest access to LAMS for one month following the workshop. This was the first time (worldwide!) that persons outside of the development project were given access to LAMS, so we are grateful to James Dalziel for this.
Discussion & questions about LAMS:
The workshop involved hands-on access to LAMS, and participants were engrossed in discovering this new tool, hence the time for discussions was somewhat limited in this session, but LAMS was discussed again in the final discussion session (link to relevant part)
James:
A regular question is whether it is possible to export LAMS sequences into VLEs?
No, not at the moment, but hopefully will be in the future with conformant systems.
It is not a trivial piece of work. At the moment can interoperate with LAMS
– share sequences etc.
Paul Hollins:
Seems to be useful for small group activities – from an authoring environment
James:
It’s a research & development project and we didn’t know what
we have today would be possible.
Bill Olivier:
From a point of view of IMS Learning Design, is one activity a combination of
an act and a service? All embedded together?
James:
That’s roughly right. Some activities embed more than one act, normally
can only synchronise between acts. So the question we need to ask is ‘does
the specification need to be changed or do we say the tool just does this?’
Each box is a combination template of role, activity, content – that makes
it easier for the teacher to use because it bundles them all up together.
We’d like to make it adaptable by users. The work is ongoing and we plan
to work on accessibility issues soon.
Implementing a Learning Scenario with IMS Learning Design- Jocelyn Manderveld, Open University of the Netherlands Educational Technology Expertise Centre (OTEC)
Jocelyn is an educational technologist with a background in educational psychology. She was closely involved in the development of EML, the EML player Edubox, and IMS LD.
The aim of this session was to provide participants with insight into what is involved in implementing a learning scenario using IMS LD, including delivery through a runtime system. A non-technical overview was provided via a presentation on the background to IMS LD from the OUNL perspective, followed by a walk-through of creating a learning activity as it is done at the OUNL, with a demo of delivery via Edubox.
A great many courses have been, and continue to be prepared and delivered in this way; most of the time involved is spent in preparing the UML modelling diagram used to plan a course delivered via IMS LD.
Discussions & questions about EML
Working online and offline; blended learning
Comment: A big problem with the examples shown is that you can’t study them offline, you need an Internet connection, to keep interactivity.
Clive Church: Strengths of this approach though, are that you can plan offline and build blended learning into the approach.
JM: The problems arise with how it’s implemented. You do need some connectivity if you want interactions.
Bill Olivier: Planning is a big part for preparing any learning experience. With authoring, a huge amount depends on authoring tools, e.g. XML Spy could be too complicated for some people. LAMS is at the other extreme where it bundles up the services etc. Online/offline is a useful choice, where students can work and buffer up the info and then connect. We want to do that with RELOAD; we have already done it with Colloquia (JISC-funded learning management and groupware system).
IMS LD and Accessibility
Pete Rainger: Have you had much call to make changes to design for individual students (e.g. those with different accessibility needs)? How difficult would it be to change?
JM: We try to build any needs into the scenario, but then we need to know the possibilities ahead of time. If a course is actually running and something happens, there are parts that you can change (e.g. educational content) and cannot (e.g. pedagogy, competencies – otherwise students lose all previous info in portfolio).
Clive Church: This is important for satisfying accessibility requirements.
Pete Rainger: Yes, e.g. with a dyslexic person, you may in this case have to build in some kind of inclusive curriculum design.
JM: You could do it and make a new ‘run’. Changing the design is a problem, but changing the content is no problem.
Bill Olivier: It all depends on how it’s compiled and run. Would be good to enable that each act can change after each run of a learning design. Would leave a gap between to allow structure to be changed. There is a distinction between global properties (those that persist across all runs) and local properties (these can change for each run)- IMS Accessibility is step one for what we need to do with these properties with regard to accessibility. There is a bigger burden on the designer in including this stuff.
Clive Church: Hopefully the tools will come out to help us. I am hopeful that there will be different types of content for different learners.
Joanna Brown: Then you could have different paths.
Bill Olivier: As users become more familiar with the environment, they may not need stuff visible all the time – there would be more scope to adapt.
JM: It’s difficult to explain all the possibilities. Using IMS LD as a base, we need to discuss functional use by teacher and student, how it looks and works. Communities of practice around Europe for IMS LD can do this, e.g. via the new UNFOLD project.
Pete Rainger: TechDIS could look at patterns of use for inclusive design. Would be a good exercise anyway, especially in the context of IMS LD.
JM: I am doing research into patterns of IMS LD, analysing courses and see if I can find 1 or 2 good IMS LD patterns.
Bill Olivier: With accessibility you can go the other way. People have to think about pedagogy so they already identify patterns of practice.
Pete Rainger: Tutors already know how to modify learning content for learners.
Bill Olivier: Surveying that would be incredibly useful.
Clive Church: What about work-based learning and motivating learners – this could be useful to re-engage people with learning.
Question: Would IMS LD be the most effective way to gather data rather than use cases? I find IMS LD quite challenging to get my head around.
Pete Raingerr: As an exercise it’s good to do.
Bill Olivier: Use cases and scenarios are good too. End users need to feed into the development of the specification, and this is one way to do it.
Clive Church: We need simple tools and to be able to play these things on any platform
The Final Session- Next Steps
The final session on the second day was devoted to discussion and feedback arising from participants’ experiences in the practical sessions and discussions. This was an important feature of the workshop- the presence of IMS LD developers and implementers alongside interested learning technologists and educators gave the opportunity for direct feedback to the specification and tool development processes.
Bill Olivier began by talking about future developments in IMS LD implementation such as RELOAD, and the new UNFOLD project which has been funded under the EU 6th Framework Programme.
Bill's slides include extensive notes and present:
- The RELOAD open source implementation project
- Collaborations
- The RELOAD editor and runtime architectures
- Animated mock-ups of possible UIs for Runtime and Editor engines, for comment and feedback
- Valkenburg Group background. Use cases demonstrated a desire for multiple tools.
- RELOAD - Nov 03 - July 04 (with possibility of funding to 05 to work on level B & C) LD. SCORM player. Partnerships important. Chose MIT open source to encourage vendors to adopt. Pen State digital image repository – investigated why people are not using repositories, found contributors want to control who their material is shared with, desire for a peer-to-peer repository like Napster but with authentication etc. Led to forming of open source partnership.
- UNFOLD. Supporting communities of practice. Standards development, trialled in practice before incorporation into standards. Fully international.
In introducing the final discussion Sarah Currier noted that the organisers were particularly interested in hearing about possible future tasks or roles for CETIS.
Using the UML Development Process to Engage Teachers and Provide Use
Cases
Most of the afternoon’s discussion centred around this idea of Rachel
Ellaway’s, while making occasional forays into other areas (which are
noted under separate headings below). Rachel noted that, while implementing
full IMS LD in XML is prohibitively expensive at present, perhaps the UML part
of the process would be useful as a kind of human-readable ‘middleware’,
to engage teachers with the IMS LD way of thinking, and to gather the ways in
which teaching is currently being done, and hence the ways in which IMS LD and
tools to support it may be developed.
Gayle Calverley added that there is also the issue of management of delivery, as well as just course design. For the former, we need to break down what is going on to work out the best way to support learning. Using IMS LD and UML as Rachel suggested would be a way of reaching common agreement on what is going on, to build for the future, getting institutions involved, through being able to show systems staff and managers what is actually required. There needs to be a link into enterprise systems, matching up what exists with what's wanted, to make cases with technical folk, to make decisions– UML diagrams could be useful as brokerage between working out what's available between technical groups and academics. It would let academics articulate course needs and let technical staff articulate system needs and capabilities. Rachel added that some things are very difficult to do just now in elearning– we need to make a distinction between what's possible and what's really difficult; this will help people understand the reality of technological implementation and help to create future paths. This ‘middleware’ would be a useful tool to enable academics to be involved, which is empowering for them as it provides a direct link into real world activity rather than as mediated by learning technologists.
Bill Olivier expressed some doubt that teachers learning UML is a viable approach, and the discussion then centred on how to create a shared expression language at a higher level than IMS LD or UML use cases, to put Rachel’s idea into action. It was generally agreed that something very accessible to enable people to take the first step, perhaps in the form of a tool, would help. Bill wondered if RELOAD could be designed to allow extra layers of interface to be developed- perhaps a LAMS type user-friendly surface with IMS LD generation underneath?
Sarah Currier asked the group about how they wanted to see these ideas taken
forward- who should generate and gather these use cases? It was generally felt
that, while different groups and institutions could create their own use cases,
there is a role for CETIS in not only gathering and disseminating them, but
also in bringing together people’s ideas about how best to express them.
CETIS would be important for making it community wide. Bill suggested that we
need to see examples of what people are thinking of, start circulating ideas
and diagrams, get feedback. There are many different approaches/ideas, e.g.
instantiations vs. generic patterns of learning design. Pete Rainger had already
noted in Jocelyn Manderveld’s session that TechDIS might be a good place
to gather use “patterns” for accessible, inclusive learning design.
It had also been noted that the UNFOLD project could do this as part of its
remit.
Discussion then moved on to specifics of expressing the use cases. Rachel suggested that we could collect ca. 30 exemplars of objects in academic use, seeking a standard format- although it may turn out that a UML diagram format is a better way of doing it. Gayle thought we could use different diagrams for academic and technical staff, different steps. From this, a structured piece of work could be put together. Bill wondered, however, about whether if we create some sort of formalism, whether it be UML, XML etc, would people be more comfortable if we had a format which allowed a lot of narrative rather than thinking in terms of a particular tool and adapting IMS LD to that tool? Gayle’s idea was that we would start with narrative, then a tool would allow conversion to a diagram, e.g. when we have to state requirements, we use a Word document which gives a first draft of a formal requirements document. This aids the human user who doesn't need a learned or constrained format. There was brief discussion about the nature of the language / terminology used in the tool, with Bill reminding us that OUNL spent 3 years developing the EML/IMS LD language, is it good enough? While this was agreed, it was noted that we want to make it an easy step-through. Dan Howell agreed that we need to enable the capturing of information in a reasonably structured format, incorporating different phases, allowing conversation, and creating a checklist of processes and pedagogies people are seeking. Bill asked whether a simplified tool with descriptive narrative for each process/step would be an appropriate halfway house between approaches.
James Dalziel noted that, when they gave LAMS to the average teacher, they expected them to instantiate into the online system what they do in practice; however, this is not what teachers describe as their experience of the system. They found that they questioned their own educational practices. They found that seeing their existing practice in a visual way can reveal a lack of reflection on aspects of their teaching practices, and this experience encourages them to challenge this. James noted that this was a surprising but very worthwhile outcome- around 50% reported that LAMS led to them reconceptualise their own teaching. So the process of working with teachers isn't just requirements gathering- the requirements are changing as they're being gathered. Bill noted that the LAMS design interface is in itself a formalism for doing that. Mark Pettigrew added that academics can't specify requirements as well as they can identify gaps in provided tools.
Reflecting on James’ comment, Phil Barker suggested that, when a learning design is mapped out, we could use simulated students and tutors and put them through the course and perhaps reach a point where the learning design broke down in practice. Or, perhaps more realistically, that it might be useful to have a toolkit for implementing learning design to encourage teachers to reflect on their practice. Grainne Conole's media toolkit (link http://www.londonmet.ac.uk/ltri/demos/media_adviser_files/media_adviser.htm) developed a few years ago contained course components, each categorised as interactive/discursive/etc.; once the tutor had built her course it would give a breakdown on what course was actually like. Perhaps something along those lines would be useful.
Bill summarised at this point by noting that it seemed that we were saying we should be working at higher level- so, how to do this? James said that we need to think about how best to represent a sequence of activities. LAMS has a level of abstraction– you can make it more simple or complex. It is aimed at teachers thinking about the flow of educational activities (ie “lesson plans”) – this was a fundamental design requirement for LAMS. Dan Howell noted that, from an FE perspective, a descriptive curriculum, learning plans, etc. translates online easily. It’s crucial to have tools for non-technical staff: if conversations don't happen, things don't happen. IMS is about supporting tools. He liked the idea of a smart, diagrammatic tool which obscures detail. The FE position is that they need to consider when to engage support and technical staff. They need to describe and decide on sequencing points at the top level before they get down to detail.
So, the process identified was that we want to describe certain things and need tools- so we need to describe process first before we can consider how to implement them. If people have any method of describing processes, Lisa Corley requested them to post to the CETIS Pedagogy Forum discussion list so we can see if learning design supports it, and see what else is needed. We need to communicate real needs, develop higher level languages. It would be very useful to have all this in one place.
Marion Manton, James, and others noted that this could be a costly and time-consuming
exercise, and that money may be needed. Lisa Corley said that JISC representatives
are coming to the next Pedagogy Forum meeting (http://www.cetis.ac.uk/members/pedagogy/articles/4thmeeting_framework),
to inform about a new JISC e-learning and pedagogy programme
(http://www.jisc.ac.uk/index.cfm?name=elearning_pedagogy)
which may cover issues which have been discussed. This programme has funded
an updated version of
Sandy Britain and Oleg Liber’s JISC report from 1999 ‘A Framework
for the Pedagogical evaluation of VLEs’ (http://www.jisc.ac.uk/index.cfm?name=project_pedagogical_vle
) The revised report which will be launched and discussed at the Pedagogy Forum
meeting in February 2004. Sandy Britain summarised by saying: there was a 1999
report on pedagogy and elearning which looked at pedagogical evaluation of VLEs.
However, the concept of VLEs seems to be breaking up– today has provided
good evidence of this– in favour of course management systems, tools,
a greater spread of granularity, architectures. Sandy is now redoing that report,
and as part of his work, is going to review some of the systems that have
been around for a while, what they do and how they do it, and what aims there
are for the future. There seems to be more emphasis on enterprise solutions
than a focus on teaching and learning and pedagogic innovation. He is keen for
people to develop learning scenarios, test them based on the framework developed,
and see how they cope, which seems to connect with what is being discussed.
The next Pedagogy Forum meeting has this same focus: would this be a useful
way of approaching the problem? Lisa suggested we carry on discussions and ideas
on the Forum mailing list ahead of the next meeting in Feb. 2004. Sandy’s
report will be out March 2004 (http://www.jisc.ac.uk/index.cfm?name=elearning_pedagogy)
, but people could bring scenarios to the meeting. Sue White asked if it was
worth doing an event for schools, to see how teachers take core material provided
and incorporate it into their own pedagogies or incorporate their own pedagogies.
Lorna suggested that people bring along examples of what they consider good pedagogy. Can good pedagogy be encapsulated by IMS LD or a tool like LAMS? Mark Pettigrew warned that he thought the term 'good pedagogy' is worrying when asking for examples of practice– teachers are deterred by the possibility of submitting something imperfect. Perhaps we should be saying ‘appropriate’ or ‘useful’ practice. Sue White noted that it’s not a case of ‘one size fits all’- variety is inherent. Mark also commented that, in practice, sometimes things you don't think will be good are very effective. James Dalziel pointed out that we can't say LAMS makes people use good pedagogy, it can be used for lousy pedagogy also! Differences in assumptions become explicit through the language used to describe it.
IMS LD and Contextual Data about Learning Objects / “Secondary”
Metadata
Phil Barker mentioned how people have been talking recently about data on how
learning objects used (also known as ‘secondary’ metadata)–
there is a need to be able to capture context in which they are used–
can we use IMS LD to describe how an object is used? Bill responded that IMS
LD describes context of use, but it is not yet clear how this can be extracted
or shared. Sarah noted that this issue had already come up in the IMS InterActive
Content SIG’s requirements gathering exercise on next-generation developments
for specs- so is worth noting again.
IMS LD and Services
Phil Barker: Regarding the concept of “services” within IMS LD –
Bill mentioned 2 services yesterday (send e-mail & conferences) –
presumably there are others? To what level do they need to be nailed down -
i.e. how precisely do we need to define how a conferencing tool should work
e.g. ‘allows threaded discussion’?
Bill Olivier: This is for future development, but it is a key issue. LAMS services
for example are very nice, but they are all within the LAMS box, they are not
yet able to pull in services from elsewhere, which is one of the key issues
for IMS LD. There were a lot of services and other features in EML but they
were cut during the development of IMS LD to keep the vendors on board; the
intent was always to reintroduce them, and this may be a pressing task for the
next stage of IMS LD. LAMS is in the lead by far on these, and the community
needs to find interoperable ways of adopting them.
IMS LD and the "concept" of Learning Design
James Dalziel noted that the IMS Learning Design specification was an initial attempt
at writing a specification based on the broader theoretical concept of "Learning Design"
(that is, "people conducting activities with resources"). Over time, the IMS LD
specification will evolve and change, and it is possible to imagine other specifications
that could implement the concepts of Learning Design, but in a way different to the IMS
LD specification. James argued that it is important not to get too hung up on the details of the
current IMS LD specification, but rather to keep discussing ideas, implementations and problems
so this helps the area evolve. In a similar way, any particular software system (such as
LAMS, EduBox or the proposed future RELOAD LD player) will implement Learning Design in
different ways, even when based on the same model or specification- this is to be expected at
this early stage of development. Over time, as a variety of systems evolve, the potential
for interoperability testing will arise, and the outcomes of this process can also be fed
back into the evolving specification and theoretical concepts.
What Does IMS LD Allow? What About On-the-Fly Changes to Designs?
Mark Pettigrew made an eloquent plea not to prohibit things that aren't expressly
supported (in the specification and also in any tools), such as nesting of activities beyond four
instances. We should make systems encapsulate zips even though they don't necessarily
know what to do with them - keep 'soft edges' - be kind to the people who want
to do what you haven’t thought of– think about implementations more
than standards. This is not just a technical issue- relates to the design of
reusable learning objects also- there should be a way to flag up if the new
author needs to deal with anything. For instance, the use of different names,
e.g. discussion forum/discussion board/conference. We need to be aware of changes,
perhaps by flagging them in the manifest when we transfer learning designs between
platforms– and the ability to put in fix ups would be very useful across
all standards.
Mark also raised the issue of run-time verification, e.g., this meeting didn’t break into the expected breakout groups, as would have been programmed into a learning design- how do we cope with that online? Signed off = signed off- except it doesn't necessarily. We have a stop point, but can we modify stuff after that stop point? Bill’s opinion was that this is a permissible concept in IMS LD, but, if a learning design is a script to get a process through an engine, how does it do that: on the fly, precompiled? Do dependencies which refer backwards/forwards restrict flexibility? James noted that there are two parts to the problem: 1. software, and 2. conceptual: to what extent do you put implementation details into learning design? In one LAMS implementation, there was a notice board before the stop point which flagged up the date of the next tutorial, which seems to have embedded context into an abstract design: how much specificity/abstraction do we put into things? What should best practice be? Lorna Campbell pointed out that this does raise the issue of how much internal contextualisation you can put in a learning object and still allow it to be reused and interoperable. This is the same problem whether using standard content or learning designs: we need guidelines for good practice in all development of educational objects.
Sarah Currier and Lisa Corley concluded the workshop with thanks to all participants and presenters for contributing to a very valuable event, and to LTScotland for their fantastic support.
Summary of Key Issues to Take Forward
Issues for Tools
- Participants should be able to work off-line where possible as well as online.
- Bill Olivier’s idea of allowing for enabling “that each act can change after each run of a learning design. Would leave a gap between to allow structure to be changed.” – This idea was in reference to the need for different designs for accessibility reasons.
- Phil Barker’s idea to develop “toolkit” for implementing learning design to encourage teachers to reflect on their practice; see Grainne Conole’s media toolkit.
- “A plea not to prohibit things that aren't expressly supported, such nesting beyond four instances - don't limit how much. Make systems encapsulate zips even though they don't necessarily know what to do with them - keep 'soft edges' - be kind to the people who want to do what you haven’t thought of – think about implementations more than standards.” (Under ‘Issues for IMS LD Specification Development’ also).
Issues for Educational Community / Course Authors and Developers
- Pete Rainger suggested that TechDIS “could look at patterns of use for inclusive design”.
- Jocelyn Manderveld suggested that, broader than this, the European elearning community needs to start documenting patterns of use: possibly the UNFOLD project’s job?
- Gathering of IMS LD use cases from UK HE/FE in the form of UML or a similar agreed expression language, as an exercise for teachers and to feed into development of IMS LD and its tools. Possibly develop RELOAD to enable this?- with an option for a LAMS-type interface with IMS LD beneath?
- Gayle Calverley’s idea: make this useful for explaining to managers and technical systems people in institutions to see what will be needed technically and with planning, identify gaps, etc.
- Possibility for CETIS to coordinate efforts.
- First: use CETIS Pedagogy Forum to share ideas for how to express these use cases, ideas, commonalities of needs, etc.: Bill “We need to see examples of what people are thinking of, start circulating ideas and diagrams, get feedback. There are many different approaches/ideas, e.g. instantiation or generic patterns of learning design.”
Issues for IMS LD Specification development
- How can IMS LD be used to extract data about how learning objects are used, a-la “secondary metadata”?
- Phil Barker: Regarding the concept of “services” within IMS LD – Bill mentioned 2 services yesterday (send e-mail & conferences) – presumably there are others? To what level do they need to be nailed down - i.e. how precisely do we need to define how conferencing tool should work e.g. ‘allows threaded discussion’? – On the agenda for future IMS LD development.
- “A plea not to prohibit things that aren't expressly supported, such nesting beyond four instances - don't limit how much. Make systems encapsulate zips even though they don't necessarily know what to do with them - keep 'soft edges' - be kind to the people who want to do what you haven’t thought of – think about implementations more than standards.
- Look at the issue of “on-the-fly” changes within a learning design.
Post-event Evaluation
The workshop was evaluated via a form distributed to all participants. The reviews were overwhelmingly glowing, and there was an emphasis on requests for future workshops of a similar nature; the need for tools, tools, and more tools; and agreement that gathering use cases, usage patterns, etc. as discussed in the meeting was a good idea.