A copy of the paper delivered at educause - 6 April 2005

Collaborating Online with an Open Source
System

 

Glen Davies, Christchurch College of Education

 

ABSTRACT

 

In 2001 the Christchurch College of
Education developed its own Learning Management System using an open source
platform. Existing e-learning platforms at the time seemed to be locked into a transmission
mode of teaching and learning, so one of the main aims of the project was to
make it easy for students and lecturers to interact and collaborate online,
based around constructivist and Vygotskian pedagogy.

 

The other key design factor was to make the
system flexible enough to be used for collaboration outside an e-learning
context. This has since led to the use of the system in a number of non e-learning
related projects and communities of practice.

 

1                      
Introduction

 

This paper will:

 

1                        
Outline the reasons for
developing an open source collaboration platform

2                        
Examine how successful the
system development project has been

 

3                        
Look at the lessons learned
along the way

4                        
Highlight some of the
institutional inhibitors/aids to making such a project successful

 

5                        
Showcase the latest
developments in the system aimed at further improving online collaboration in
an e-learning context 

 

 

2                      
Discussion

 

2.1                 
Background to the Project

 

In 2001
the Christchurch College of Education embarked on an evaluation of Learning
Management Systems in order to support a growing number of courses with online
components, as well as to supplement face-to-face delivery. A shortlist of
three currently available commercial systems was arrived at through a review of
evaluations undertaken by other institutions internationally. Copies of the
three systems were obtained and installed for testing. The testing consisted of
both the setting up of fictitious courses, and also the running of several live
courses with College students and staff. Although all three systems provided a
reasonable amount of functionality, feedback from lecturing staff involved
highlighted some key deficiencies:

 

·       All systems provided areas for
adding content, and areas for online communication/collaboration, but the two
were in quite separate areas, with no easy way of blending the two. Students
had to read content in one place, and discuss it in another.

·       The discussion/collaboration
areas provided very few tools for lecturers to manage the large volume of
postings generated by the discourse based learning pedagogies being used.

·       Set system structures gave very
little flexibility for lecturers to construct courses in the they wanted.

·       There was very little scope for
students to contribute content other than forum postings and assignment
dropboxes.

·       The systems were very much
designed around a transmission mode of teaching.

 

As well
as these limitations the systems licensing structures were expensive given the
small number of courses that would initially be using the system.

 

Along
with the three commercial applications tested, one course was run on the open
source platform LearnLoop (http://www.learnloop.org) in semester one 2001.
Although the lecturing staff involved were impressed with the flexibility of
the system, the project seemed to have stalled at the time, and the frames-based
architecture limited some of the system potential, so it was decided not to
pursue this system further. The basic LearnLoop system design did however
influence the design of the in-house system.

 

Taking
these considerations into account it was decided to develop a small in-house
system. The key requirements of this system were:

 

·       Ease of use for lecturers, who in
many cases had only basic ICT skills

·       The ability to easily mix content
and collaboration

·       To allow flexibility in course
design with no pre-imposed structure

·       Tools for managing large amounts
of online discussion

·       Ability for students to add
content

 

Lecturers
also wanted the ability to move online learning into a more social
constructivist pedagogy, viewing learning as a social process, and seeing that
meaningful learning occurs when individuals are engaged in social activities (Kim,
2001). The system therefore needed to provide tools for students to engage in
social interaction with their peers as well as lecturing staff.

 

As well
as the delivery of online courses there was also an interest in using the
system outside the traditional ‘course’ context for the development of
communities of practice, “groups of people who share a concern or a
passion for something they do and who interact regularly to learn how to do it
better” (Wenger).

 

2.2                 
System design and development

 

The
first decision that needed to be made was the development platform to be used.
Although there were some desktop client-based e-learning platforms in existence
at the time it was decided to develop a fully web-based platform in order to
allow easy support of multiple client platforms. A php/mysql/apache development
environment was chosen, largely due to the existing skill base with these
development tools, but also in order to make the open source release of any
completed system possible.

 

An agile
development methodology was adopted because of the short timeframes involved in
providing a production system. This choice was also influenced by a belief that
iterative development involving tight development and release cycles, and a
close relationship between users and programmers, would  be needed in order to keep pace with the fast
changing area of e-learning while still providing stable and usable software.

 

System
development was begun in September 2001, and a beta test version was made
available to lecturers in November. Initial testing proved successful and the
system went live in time for the 2002 academic year.

 

At the
end of 2002 it was decided to do some further development on the system, based
on the experience and feedback from a year of use. There had also been some
interest in the system from external parties so it was decided to release the
whole system freely under an open source GNU GP License (http://www.gnu.org/) with
the name Interact. Since the middle of 2002 the system has been available via
the Sourceforge open source project site under the name Interact, and has more recently
moved to its own project site at http://www.interactlms.org/

 

2.3                 
Key system differences

The key thing that differentiates Interact
from most existing proprietary
systems is the ability for lecturers to start with a blank slate and construct
their e-learning environment in any way they want, with the ability to easily
mix content and interaction in the same space. This was one of the key aims of
the development team, and it was pleasing to see that it was this aspect of the
system that attracted academics from other institutions, both in New Zealand
and overseas, to the system.

 

Over the past 7 years, I have had
considerable experience with commercial and open source e-learning systems and
continue to explore new ones as they come to my attention. I initially came
across Interact in the Spring of 2003 and after spending a few hours exploring
it came to realize that it had some remarkable features and a great deal of
potential for further development. Since that time I have used Interact to
support my own teaching and plan to continue to use it in the near future.

A simple
inventory of the features of the Interact software, including those scheduled
for development in the short term, demonstrates that it is similar to or better
than other well known products that are available to educational institutions. What
really distinguishes Interact, however, is the way in which those features can
be used in designing learning experiences for learners. Interact has a high
degree of pedagogic flexibility in that it allows instructional designers a
great deal of latitude as they design a learning experience for students.
The options available to instructional designers allow them to easily include
learning activities that educational research has consistently shown to positively
contribute to learning, such as group work, journal writing, interaction, and
feedback management.

Dr Gary Hepburn, Assistant Professor
Acadia University
School of Education
Canada

 

2.4                 
Success of the Project

The relative success of the project has been
measured in several ways

 

·       Feedback from lecturing staff on the systems flexibility and ease of
use has been mostly positive.

·       The system has provided a stable online learning and collaboration
platform for staff and students at the College for the last three years with at
most one hour of down time.

·       There has been a considerable amount of interest in the system, both
nationally and internationally. It is now in use in tertiary institutions in Canada, Washington,
and Brazil,
to name a few, and is also used for a number of Ministry of Education community
of practice related projects.

·       Overall the cost of the system to the College has been similar to
purchasing a commercial product, and there is the potential for the costs to be
offset by growing revenue streams from external use of the system.

 

2.5                 
Lessons Learned

There are several key lessons that have
been learnt by the development team over the course of the project:

 

• Regardless of the expected life of the system always plan for
  scalability and extensibility. When initially started it was envisaged
  that the system would be a temporary measure while other commercial
  offerings matured. This led to a fast development timeframe, and a lack of
  modularity in the core system structure. Once it was decided to continue
  with the system this led to an extensive amount of refactoring of the core
  system in order to modularise the code and make it more scalable and
  extensible.

 

• Adhere to the open source mantra of ‘Release early, release
  often’ (Raymond, 2000). The system should have been released open source
  the moment development started. On reflection the 18 months that lapsed
  between the start of development and release of the open source project
  would have provided more opportunity for the system to gain traction with
  other users both nationally and internationally.

 

• Use Agile development methodologies (http://agilemanifesto.org/).
  Advances in elearning and web applications are moving too quickly for
  traditional development methodologies to keep pace.

   

2.6                 
institutional inhibitors/aids
to the Project

The main institutional aid to making this
project a success was the willingness of management to trust the staff involved
in the project, to accept their decisions, and to allow them to take calculated
risks. With only minimal requirements for project briefs and business cases in
order to gain approval, the project was able to quickly move forward. This of
course was helped by the very low capital costs involved in the project.

 

The only inhibitors were the unwillingness
of some teaching staff to accept the system as just another tool to use in
their course delivery, instead choosing to see it as a threat and a ploy from
management to increase workload and reduce staff numbers. The other factor was
small battles over who was allowed to maintain which parts of the system, with
much time spent discussing issues of academic freedom versus centralised
quality control of content and design.

 

2.7                 
Latest Developments

Now that the core functionality of the
system is in place the project team is working on expanding the social
interaction aspects. There are many features that can be added in order to
facilitate this. One thing that most e-learning platforms do is restrict users
to seeing only those users in their current courses. The latest system
developments will allow the setting up of user profiles that will match
learners with those with similar interests that may not necessarily be in the
same course or intake. It is hoped that in working with these concepts and the
ideas around communities of practice that new pedagogies for online learning,
that break out of the traditional teacher, students, course moulds can be
development and researched.

 

3                      
Conclusions

 

The College of Education took a risk in deciding to
develop its own in-house e-learning and collaboration platform. In most cases,
developing software is an expensive process and can be fraught with problems.
However in this case the risk has paid off. The College has a stable, scalable
platform which is not only cost effective but provides a potential income
stream. It also gives the College the flexibility to research and implement new
methodologies in online learning and collaboration, something that a commercial
proprietary system would not have afforded.

 

 

4                      
Acknowledgements 

 

Gordon Findlay – College IT Manager at time of
project commencement

Derek Chirnside – Educational Designer

Neil Helson – Current College
IT Manager

Maureen Doherty – Director Learning and
Information Services

 

5                      
References 

 

Kim, B. (2001). Social constructivism. In M. Orey (Ed.), Emerging perspectives on learning, teaching,
and technology. Retrieved November 30, 
2004 , from http://www.coe.uga.edu/epltt/SocialConstructivism.htm

 

Wenger, E. Communities of
practice: A brief introduction. Retrieved November 30, 2004, from http://www.ewenger.com/theory/index.htm

 

Raymond, Eric Stephen (2000). The
Cathedral and the Bazaar.  Retrieved
November 30, 2004, from http://www.catb.org/~esr/writings/cathedral-bazaar/cathedral-bazaar/