The Iconex project  at the University of Hull was funded under the JISC's 5/99 Programme to demonstrate the value of small, portable, pieces of digital content in assisting student learning. The project is creating a repository of exemplar interactive Learning Objects, many of which are already available for use and reuse. This repository is intended to stimulate cross-fertilisation between disciplines to develop generic views of types of interaction, and to encourage the reuse of Learning Objects.
In this paper, we explore some of the more wide-ranging issues which have arisen during the project, and attempt to demonstrate why consideration of Learning Objects and their role is relevant to all readers of Ariadne, not just to Learning Technologists.
What is a Learning Object?
- Learning Object
- any entity, digital or non-digital, which can be used, re-used or referenced during technology supported learning
The IEEE definition of a Learning Object is, perhaps deliberately, somewhat broad in scope, and could quite easily be applied to everything around us. The definition is therefore helpful in reminding us that we continue to learn throughout our lives, and that every interaction we undertake is a potential learning experience, but less helpful when trying to tie down some concepts in order to progress discussion in this area.
The Iconex project worked with a narrower, non-exclusive, conceptualisation of the Learning Object, specifically considering three related categories of object, all of which are digital;
- Generating Objects (Objects which produce interactions)
- Connective Objects (Objects which can be connected together to produce richer interactions)
- Adaptive Objects (providing for example, enhanced accessibility where appropriate by accessing a student profile)
Each of these is defined in terms of their behaviour with respect to the learner, and there is a presumption that these Objects become progressively more complex to create, more Interoperable, and of greater value to both Learner and Teacher as we move from Generating to Adaptive. Passive Objects (those which simply deliver content of some form to a non-participating Learner) are not a specific focus within the Iconex framework, although they do have a continuing, and significant, role to play in the learning process.
Figure 1: A working example of a Generating Object from the Iconex Repository [connection lost -Ed.], created in Flash
For readers able to view Flash objects*, Figure 1 illustrates a simple Generating Object, drawn from the Repository. It was designed by educators in UK Further Education for use in the area of Basic and Key skills. As with other objects in the Repository, the generating object represents something of a work in progress. The current development version of the object is, for example, perhaps overly enthusiastic in providing immediate feedback on your work before you've finished! Nevertheless, it serves to illustrate a very simple tool for reinforcing Basic and Key Skills, and which is suitable for embedding by the teacher in a variety of learning contexts.
Learning Objects such as this have traditionally been expensive to produce. They are typically defined and specified by educators, with this domain specific expertise then translated into software and polished by teams of multimedia developers. Although such developers re-use code from object to object, the production of even small interactive learning objects is labour intensive. The larger the object, the greater the expense involved. Some such objects have in the past formed almost entire systems delivered on CD-ROM. In producing exemplar objects, Iconex has taken a distinctive approach, working with educators to develop specifications which ideally may be used in more than one context, and delivering implementations where key informational content is abstracted from the object, and provided in an external data (in fact, XML) file. This approach has several advantages.
- By obtaining information from an external data file, the object becomes effectively a rules-based rendering engine, and can provide a variety of instances by referencing different data files, without a need to re-write the object multiple times. In this context re-usability becomes a very practical proposition.
- The object can store user information — such as results — in an external data file.
- In addition to presenting different information, the object can reference external data files which modify its presentational form. The same object can therefore be used to demonstrate its concept in a different language, or in an optimal format for those with disabilities.
- Such objects tend to be small, focussed, and easy to deliver over the web. The fine granularity of such objects acts as a further stimulus to re-use.
Such objects are clearly not designed primarily to stand alone (although there is nothing to prevent an educator using them in this manner if so desired), but to be integrated with what have become commonly termed Virtual Learning Environments (VLE). A Flash object such as the above may of course be played from any web page, whether within a VLE or not. There are problems, however, of a large and small nature in establishing meaningful placement of such objects into VLEs. Many of the first generation VLEs currently in use were developed whilst specifications and standards were little more than drafts. They have a tendency to be monolithic and resemble “black box” solutions — difficult both to integrate into broader systems, and to integrate externally generated Learning Objects within. This difficulty is compounded by the tendency of the first raft of relevant specifications and standards to be data centred, rather than focussed on how these data might usefully be connected (although we had to start somewhere!). For this reason, consortia such as the Open Knowledge Initiative  are currently focussed on defining application programming interfaces, or APIs, which are designed to enable systems to communicate with one another in standard ways. The OKI are collaborating with IMS  to push specification and standards development in this direction. When learning environments not only store data, but transmit those data in standard ways, the meaningful integration of interactive learning objects becomes at least possible.
There is clearly a large amount of work to undertake before that potential becomes a reality. A key element of the Iconex project has been the engagement of project staff with significant areas of the community; whether those involved in providing and packaging content, those developing VLEs, or those involved in the specifications and standards process. This engagement with the community is central to both anchoring and driving the development of learning objects. Collaboration, and collaborative frameworks are also crucial to success. Iconex is in the early stages of dialogue with the US-based Merlot Project . The projects have much in common, with some interesting differences of emphasis. Merlot aims fundamentally to establish a learning object (not necessarily interactive learning object) "community of use", and has been successful in attracting over 7,000 objects and nearly 9,000 registered community members (plus many thousands more who benefit from using the objects) in the five or so years of its development. Such communities of development and use — and the overlap is significant — provide the potential for interdisciplinary communication and cross-fertilisation, and a context within which to generalise developments. The nurturing of such communities in the UK is an obvious next step in further aiding the development and take-up of learning objects, and Iconex is keen to learn from the Merlot experience and facilitate similar growth here.
It is significant also that these developments come at a time when university communities are drawing together systems, processes and expertise to create institutional portals — which are rapidly becoming viewed as the essential step in providing coherent experience for both off and online learners and staff. It is important that in joining up institutional systems to provide what is fundamentally a horizontal portal, this portal does not then become a portcullis, and developments in industry such as Sun's Open Net Environment  (encompassing portal, learning environment, and more) are of obvious relevance here. We must retain the perspective that online expressions of communities of use — whether regarding LOs or other vertical issues — can be integrated into this development as seamlessly as possible, and ensure that the product we offer reflects the real needs of a real community, rather than a whim of the developer.
- The Iconex Home page is at: http://www.iconex.hull.ac.uk/
- The Open Knowledge Initiative Home page is at: http://web.mit.edu/oki/
- The IMS Home page is at: http://www.imsglobal.org/
- The Merlot Home page is at: http://www.merlot.org/
- The Sun ONE Home page is at: http://wwws.sun.com/software/sunone/
Head, Academic Services Interactive Media
University of Hull