May 14, 2004
A New VUE of Digital Content – The Virtual Understanding Environment
Phil Long
Teaching in a digital environment has brought the traditional hierarchical classroom structure to the on-line world, and a range of tools has emerged to help faculty transfer their course materials and their teaching structure onto the web. At MIT we have carved two domains, distinguished by audience, for tools that support online learning. For the internal audience, MIT students and faculty, the day-to-day give and take of teaching and learning is centered on the classroom and is supported by the Stellar learning management system. For the external audience, beyond the halls of MIT and extending around the world, MIT course content is seen through the window of the OpenCourseWare (OCW) publishing system. Both of these systems strive to recreate the familiar structured organization of traditional course materials, and both do it well.
However, new projects are attempting to look beyond traditional ways of presenting and viewing online educational materials. The critical thinking and creative knowledge building required to support educational goals requires a flexibility and responsiveness that may differ by discipline and pedagogy. For instance, at MIT the MetaMedia project provides a mechanism for students and faculty to select, arrange, annotate, and present multi-media objects to convey a new narrative approach to scholarly argument. A different approach is being pursued at Tufts University in the Visual Understanding Environment (VUE) project, which is a desktop software application developed with support from the Andrew W. Mellon Foundation. The VUE project explores the use of concept mapping not only to graphically represent ideas, processes, and their interconnection, but also to manipulate the content of the information being mapped.
View imageFigure 1. A concept map of concept maps. (1)
Concept Maps for Education
In educational settings concept maps are used to
- organize instructional materials for individual courses or entire curricula,
- serve as navigational aids for hypermedia,
- scaffold understanding,
- consolidate educational experiences,
- improve affective conditions for learning,
- aid or provide an alternative to traditional writing,
- teach critical thinking.
Tufts University’s VUE uses a highly flexible, visual interface for mapping, structuring and semantically connecting electronic content drawn from local file systems, the WWW or FEDORA-based digital library systems. Unlike the concept mapping tools described above, VUE goes beyond creating flat representations of information and becomes an interactive interface to manipulating the data itself. VUE is not, however, a replacement for course management systems but an application designed to address the activity of structuring and presenting digital materials and the ideas that they contain during learning and instructional processes. The maps created with VUE could become another resource within a course management system or digital repository, literally offering students and instructors an alternative view of a set of curricular materials. VUE leverages the open and extensible development platform of the Open Knowledge Initiative (OKI) and the FEDORA digital repository to develop a visual environment for structuring.
View imageFigure 2: Dimensions and consequences of biotechnology. Each node on this map represents a category and is linked to a relevant local or remote digital resource.
VUE will provide students with a flexible utility for structuring and restructuring resource-based curricular maps, as well as a tool to help them justify their arguments, explanations, and conceptions by linking to digital sources. These concept/content maps can then be shared with instructors and peers for further review and discussion. The explicit act of organizing, annotating, and connecting educational resources via a concept map is a powerful educational exercise. Concept maps engage students as active participants in the structuring of information into knowledge and meaning. Through establishing personal connections between materials and/or by adding additional resources to an existing instructional concept map, students begin to relate new information to preexisting knowledge.
Concept maps are used at the Tufts University School of Medicine by teams of students within the Problem-Based Learning curriculum to demonstrate their understanding (and gaps in understanding) of clinical conditions and solutions. These maps traditionally have been done on white boards during class sessions and have been found to be a pedagogically effective tool. Maps created by students early in the class can alert faculty to pre-existing misconceptions or incompleteness of a student’s understanding. Post-instruction program maps can help instructors assess changes in student understandings, providing a means of assessing learning.
As valuable as concept maps have been they are relatively static, representing interrelationships among ideas that, even when created using typical software applications can at best include URLs to link ideas to web resources. When concept maps and the ideas they represent are connected to digital resources themselves rather than just their links concept maps also become content maps and this is what VUE does.
VUE extends the power derived from graphical representation of knowledge into a tool for mapping against, and drawing from, persistent collections of content contained within digital libraries. VUE is also an integrated digital library application. Using OKI’s Digital Repository Open Source Interface Definition (OSID), among others, VUE allows users to search, browse, retrieve content from digital repositories and upload resources into these systems. The FEDORA digital repository was used in the initial implementation of VUE. However, VUE can be used along with other digital repositories such as DSpace, ArtSTOR, JSTOR or any repository that implements and expose an OKI interface.
Tools like Inspiration allow one to visually drag a node on top of another, but the effect is only visual. Complex, nested hierarchies of topic elements and materials are not fully supported. Maps in VUE are represented as XML structures that can be parsed by external applications and rendered with alternate style sheets or graphic programs. This enables maps, with all of their semantic connections, to be rendered in several different ways – textual or outline mode, 3D, etc. Version 1.0 of VUE provides an outline view of the base maps.
VUE Technical Design - Application Architecture
The VUE client is built in an object-oriented, modular fashion to preserve the logical separation between the application’s core subsystems and user-generated data. This architecture provides building blocks on which future functionality will be built. VUE 1.0 consists of the following high-level components:
A VUE map is, in essence, a complex node containing a collection of topic, content, and relationship nodes along with their descriptive metadata. It is represented as an XML data file separate from its presentation characteristics. This allows user generated semantic structures to be parsed by applications outside of VUE. These separate files are managed by the application file subsystem. In addition, a map contains references to style sheets used for data visualization. As the map grows in complexity with thousands of nodes and relationships, a very robust XML schema is necessary to efficiently capture the structure and semantics of the information. The physical properties of the VUE map is represented using the Scalable Vector Graphics XML standard while the Metadata Encoding and Transmission Standard (METS) is used to capture the administrative and descriptive metadata for the all nodes. The METS schema is a standard for encoding descriptive, administrative, and structural metadata regarding digital objects.
The learning template module allows instructors to customize and/or constrain what types of nodes and links a student may create. This feature is extremely useful for concept mapping exercises designed to scaffold critical thinking and problem solving. Rather than constructing a map freely, a student may be required to first generate a hypothesis, then gather supporting evidence and post a final solution. Two templates are provided in VUE 1.0. A default, unlimited template will enable all of the application’s core functionality and a pathways template will allow users to create sequenced pathways through resources contained on VUE maps. VUE 1.0 also supports two types of visualization: a two-dimensional map containing primitive shapes and links, and a hierarchical list. However, the visualization subsystem supports the future integration of additional rendering engines. For example, a 3-D viewer might permit a user to “fly” through a map containing a large set of nodes.
Digital Repository Management
The repository management subsystem is the heart of VUE, as it provides interaction with persistent collections of digital materials, whether they are found on local or remote file systems. As with Authentication and Authorization, this subsystem was built from OKI OSIDs. VUE implements the File and Digital Repository OKI OSIDs to support communication between the VUE, file systems and digital repositories.
Basing VUE on a judicious selection of established and emerging open specifications allows other institutions to incorporate VUE into their environments more easily and provides a broader user base that can inform future versions of the application.
View imageFigure 3. Welcome to VUE – the introductory page with documentation and getting started assistance.
View imageFigure 4. The MIT’s Computationally Intensive Computing Community of Practice website represented as a concept map, but with the actual contents of the site included.
For comparison, go to the actual site. In Figure 4, note that the left pane provides a range of search capabilities across an expandable set of OKI exposed digital repositories. In this case these range from a set of learning object repositories in Canada, OCW web pages, Tufts Fedora Digital Repository, the local iTunes and iPhoto repositories on the OS X G4 on which the software is installed, and local file system of the hard drive. Moving content into the Concept map is a drag and drop process.
The Future VUE
VUE will be piloted in several Tufts’ classes in Fall 2004. There is also interest in integrating VUE with the SAKAI Project tools. SAKAI is a grant-funded, collaborative development project between MIT, Stanford, University of Michigan, Indiana University, the uPortal Consortium, and the Open Knowledge Initiative (OKI), to create a set of open source, web-based, modular, learning management tools. Tufts is a member of SAKAI’s Educational Partner Program (SEPP) and is naturally interested in having VUE work with the other SAKAI tools. To download a copy of VUE to try out, go to http://vue.tccs.tufts.edu/References:
1. J. W. Coffey, M. J. Carnot, P. J. Feltovich, J. Feltovich, R. R. Hoffman, A. J. Cañas, J. D. Novak (2003). Summary of Literature Pertaining to the Use of Concept mapping Techniques and Technologies for Education and Performance Support, Technical Report submitted to the Chief of Naval Education and Training, Pensacola, FL.(PDF)
2. Jonassen, D. (2000). Computer as Mindtools for Schools: Engaging Critical Thinking. New York: Prentice Hall.
3. Novak, J. D., & Gowin, D. B. (1984). Learning How to Learn. New York and Cambridge, UK: Cambridge University Press.
4. Novak, J. D., & Musonda, D. (1991). A twelve-year longitudinal study of science concept learning, American Educational Research Journal, 28(1), 117-153.
5. Schau, C. & Mattern, N. (1997). Use of map techniques in teaching statistics courses. The American Statistician, 51 (2): 171-175.
TrackBack
