<http://www.skeptron.uu.se/broady/dl/moca-proposal-march2002.htm>
Extract from
Wolfgang Nejdl et al, Personalized Access to Distributed Learning
Repositories (PADLR). Final Proposal, March 25, 2001, pp. 9-12.
4 Module: Server and Client Side Tools
4.1 Modular Content Archives
Contributing Research Groups and PIs. IfN Braunschweig (Ulrich Reimers), KBS
Hannover (Wolfgang Nejdl), Uppsala (Broady), CID (Broady)
Working Title. Archives: Intuitive Storage, Use and Retrieval of Archived
Educational
Media.
Problem Description. At every university usually plenty of material is
available or being produced for the planning, preparation and execution of
curricular activities like lectures, seminars and project work. This material is
an inhomogeneous amount of content of various type. These could be lecture
slides, presentations, scientific graphics and texts, description of
experiments, software, animations, simulations.
Existing distributed learning environments often demand content to be molded
into proprietary and application dependant formats. Therefore it is difficult to
re-use content for different purposes and different audiences. Hence, teachers
today are prone to produce archives and curricula based on redundant
information, proprietary applications and formats, and non-modularized
solutions.
An intelligent and flexible archiving, management, allocation and distribution
of this modular content is an intense problem. Several commercial products like
Hyperwave or Lotus offer certain functionality providing a step into the right
direction but there is still the lack of a sophisticated archive system to
optimally support curricular activities. As mentioned earlier in this proposal a
central server approach is due to the distributed nature not the optimum type of
architecture, and - building on the Edutella infrastructure - one of the
distinguishing feature to other projects (e.g. the German “Teachware-on-Demand”
project) is exactly this distributed peer-to-peer infrastructure, with plugins
for different kinds of peers participating within the Edutella network.
Intelligent archiving and flexible and intuitive access to modularized content
come along with several problems: Content modules are often detached from
context so that interrelated content modules can hardly be found and are
therefore of no use. This for instance will prevent an instructor to answer a
students question with appropriate instructive material if this material is not
intended for that particular course. Content modules can be of different file
types. Locally distributed creation leads to the need for version lists. Of-
fline archiving of created content modules is difficult and tedious. Instructors
have no fast access to content modules due to the lack of appropriate search
strategies and missing context information. Students have no access to many
interesting content modules. Especially colleagues from Sweden and Germany will
produce same content in different languages that should be linked.
Another problem is that in many educational settings teachers and students are
not able to profit from the international development of agreed markup schemes
that are evolving within research communities (mainly SGML-based or XML-based,
such as the TEI encoding guidelines among scholars in the humanities). Since the
tools and practices used in courses often do not keep pace with such
international de facto standards, the wellstructured content of existing and
emerging digital research archives are in many cases not easily available to
teachers and students.
Yet another problem concerns the rapidly increasing volume of scientific
results/ deliverables available at the WWW. In some rapidly evolving fields,
such as bioinformatics, there are hardly any course books. This means that
curricular development will be dependent on teachers’ ability to overview,
navigate, identify and re-use appropriate selections of existing digital
archives. There is a need for better principles for the design of modularized
content repositories.
There are also a series of more narrow technical problems. Since existing
materials are often stored in various file formats and developer versions along
with non-valid filenames, it is difficult and sometimes impossible to re-use
information.
Research Plan and Deliverables To solve the described problems the
following tasks have to be accomplished:
Metadata and Windows Applications. A strong focus lies on the integration
of (Windows) de-facto standard file formats likeWord, PowerPoint, JPG, GIF,
HTML, Mathcad, Windows executables etc. into Edutella archives. We will
investigate, which kind of constraints this leads to, how Edutella metadata are
used especially in this environment.
To access this content very fast efficient and flexible text/keyword search
facilities are necessary. One situation that occurs during a lecture is the
student having a question that deals with topic outside the primary scope of the
lecture. Using these sophisticated search facilities instructors can access
appropriate content very fast and answer the students question instructively.
Another important aspect is context information for content modules. If every
single module can have appended context information, it will be possible to
provide fast access to connected modules via visual presentations (context
maps).
The described tasks will lead to intensive cooperation with the ”Edutella” and
the ”Personalized Interfaces” modules. (IfN)
Repository Architecture and Tools. We propose a local database system
with a client server architecture to support instructors in an optimum way. This
single database functions as a university wide repository that can exchange
metadata and content with other distributed repositories as described in the
”Edutella” module. Commercial products like Hyperwave will be examined if these
can be extended to suit our needs. XML/SOAP et al can function as a standardized
protocol for the exchange of metadata and data between repositories or a
repository and a user. Similar developments at other institutions will also be
taken into account (for previous work see e.g. [27]).
An important issue is the linking of personalized electronic student portfolios
with such a repository. Students will be able to download interesting material
to their own electronic portfolio and manage their content. These issues will be
discussed in cooperation with members of the METAFOLIO group. (IfN)
Another task is to develop and explore methods and tools for the design and
creation of modularized content archives, as well as develop and explore methods
and tools that allow teachers and students to access and use already existing
repositories on the web, with a strong focus on modularity and re-usability in
different context, based on the content archive systems described in the
previous paragraphs. They should support the teachers and students to work with
these archives. For a specific course the teacher might propose the students
certain paths through the archive and certain subsets of content modules to be
used by the students and in some cases added to their portfolios. (IfN, KBS,
CID, Uppsala)
The knowledge management tool Conzilla already supports the organization of
annotated content into personalized portfolios in a way that is compliant with
the emerging standards for automated information exchange (XML) and metadata
handling (IMS). We will modify Conzilla to enhance this support in various ways,
based on the structure of this repository, and on user-feedback from our
activities in other modules. (CID)
Metadata-Based Learning Repositories. Based on the Edutella submodule we
will implement a second server implementation called an open learning repository
(OLR), which just stores (RDF) metadata (both for classification/annotation and
for structure) in a central database, but no content (which is accessible via
URLs/HTTP).
Resources potentially distributed all over the Internet can be combined to
entities we call “Courses”, logical documents integrating content from different
sources. While the content stays at its original location our database holds all
information about the structure of each course and metadata about the content.
The responsibility for maintaining content stays with the people providing it.
When browsing a course all HTML-pages are generated dynamically.
The system provides a platform for testing out different navigation schemes
including traditional concepts such as hierarchical tree structures and
tutorial-like course-trails as well as highly adaptive approaches (personalized
navigation) – see the personalized access submodules. (KBS)
A second system will explore an alternative architecture based on XML, XML
Schema and JSP, again compatible with the Edutella functionalities. It will
include a WebDAV-based courseware authoring module which enables geographically
dispersed courseware authors to collaboratively work on the course contents, and
a standalone, JSP-based courseware publishing engine which can achieve flexible,
dynamic courseware presentation. The system can be aware of any changes of the
course contents and automatically reflect the changes on the Web at any time,
again customizable by metadata. (KBS, CID)
Dissemination, Testbeds and Evaluation Dissemination of the achieved
results will be accomplished by scientific publications in appropriate journals
and by demonstrations at symposia.
In several testbeds we will show the impact of the developments in this module
on teaching and learning. In Braunschweig, we will use several courses in the
field of communications, encouraging exchange to other interested faculties from
other universities, in Hannover we will use several computer science courses and
the ULI project. At Uppsala University several courses in humanties and social
sciences and bioinformatics will provide suitable testbeds.
Collaboration and Scholarly Exchange Interactions with other modules:
1. Edutella module (Exchange Facilities / Basic Infrastructure)
2. Automatic extraction of metadata and ontological information
3. Personal Search Engine
There are also close links between student portfolios and content archives.
The students should be offered (and themselves able to create) content in
portable modularized formats, suitable to be incorporated into their portfolios
and to be re-used for various purposes that may not be foreseen by the teacher.
Therefore the development of portfolio practices presupposes the creation of
archives of content modules.
Use research visits (2 weeks up to 3 months) (Braunschweig, Hannover, Stanford,
Stockholm, Uppsala) in order to integrate design and development within this
module and with other modules. Workshops will be suitable to promote
collaboration as well.
URL of this page is
http://www.skeptron.uu.se/broady/dl/moca-proposal-march2002.htm>
This HTML page created by D. Broady
Last updated 2011-03-09