Skip to content.
Sections
Personal tools
You are here: Home » Features » Tackling Sustainability

Tackling Sustainability

Christina Smart
Last modified 15 May, 2006
Published 15 May, 2006
For small open source development projects sustainability is always an issue. Project teams must constantly worry about where the next phase of funding will come from to continue developments. This article looks at some of the sustainability issues for technical development projects in the e-Learning Programme, illustrated with discussions from the recent OSS Watch conference, Open Source and Sustainability 2006, and looks at what JISC is doing to address these critical issues.

Technical developments in the e-Learning Programme have been focussed around the e-Framework for Education and Research. The e-Framework is a high level, international initiative to “facilitate technical interoperability within and across education and research through improved strategic planning and implementation processes”. The work covers areas of e-research, e-administration as well as e-learning and is to some extent a top down strategic initiative [1].

At the other end of the scale, in the UK the JISC is also funding small technical projects to develop web service toolkits in several of the e-learning domain areas of the e-framework: what could be considered a developer-based, bottom-up approach [2].

Building on the work of these Toolkit projects, Reference Model projects in several domain areas, such as assessment and e-portfolios have been considering how web services might work together to solve particular business problems in their areas [3].

So where does the top down strategic approach meet that of the bottom up technical developments? The answer is in educational institutions which lie in the middle of these two approaches. The idea is that institutions will eventually be using reference models and combining toolkits to address their own unique needs, for them the sustainability of the toolkits and the e-Framework initiative will be a key factor in whether they adopt this approach.

Sustaining ideas

In response to the outcomes from previous programmes the technical developments in the e-Learning Programme have been designed to be much shorter and more focused. Each of the web service toolkit projects lasts for just six months. This allows the results of the projects to be fed back quickly to the programme team and to be fed into the demonstrator phases of the programme. Scott Wilson describes how this works;

“There is a fat and thin plan. The toolkit phase is the thin phase, with lots of small projects that are doing very particular pieces of work and are very focused. Following that is a fat phase which is the demonstrator projects. These projects use a whole bunch of these toolkits within an MLE in their institutions, so they are broad but shallow – where the toolkits are narrow but deep. And these two phases alternate across the lifetime of the programme. So the demonstrator projects identify problems and gaps which then feed into the next phase of toolkit development projects.”

While short six month projects have the benefits that the results are quickly fed back to the community, there is a disadvantage that developers tend to move on to new projects at the end and are of course highly employable! This is a problem for new projects which are developing the toolkits further and represents difficulties for the sustainability of the ideas, expertise and code. JISC have started to address this issue by ensuring that new toolkit and demonstrator projects have funding to pay former toolkit developers some consultancy. Tish Roberts, Director of the e-Learning Programme explains the rationale behind this approach:

"We recognise that sustainability is a big issue for the short term toolkit and demonstrator projects. What we don't want to do is lose all the expertise that developers and project teams have built up during their projects. That's why the current set of demonstrator projects have funding to work with the developers of the original toolkits. This should address the survival of the knowledge."

Sustaining communities

A more challenging aspect to sustainability is how the toolkits and e-framework will be used by educational institutions. A key step on the path to widespread adoption of open source projects is developing active user communities, as Stephen Downes points out in his paper: “Models for Sustainable Open Educational Resources.”[4]:

“Though there is great temptation to depict the sustainability of Open Educational Resources (OERs) in terms of funding models, technical models or even content models - and no shortage of recommendations regarding how each of these should proceed - it seems evident that any number of such models can be successful. But at the same time, it also seems clear that the sustainability of OERs - in a fashion that renders then at once both affordable and usable - requires that we think of OERs as only part of a larger picture, one that includes volunteers and incentives, community and partnerships, co-production and sharing, distributed management and control.”

Developing an active user community has contributed to the success of many open source projects. The Moodle open source content management system project team has paid particular attention to developing an active user community by using discussion fora and user conferences (Moodlemoots) and as a result now has over 100,000 registered users [5].

Open Source and Sustainability

Issues surrounding developing funding models and open source communities were discussed at the OSS Watch conference, Open Source and Sustainability 2006 in Oxford at the beginning of April [6].

Funding Models

LAMS (Learning Activity Management System) is a high profile open source e-learning system which has also put lots of effort into developing its user community, with users in 82 countries world wide. At the conference, James Dalziel, its chief architect described the path that LAMS had taken to get to its current position. James said there had been three key decision points for the LAMS project, firstly to get initial funding to develop a piece of software to address a clear gap in the market. Secondly, the decision to go open source and thirdly to completely re-engineer LAMS to ensure its scalability for the future. At each point funding was required to develop the project – and ultimately in the case of LAMS this was provided by James’ own institution Macquarie University in Sydney [7].

On the basis of the large user community LAMS International is funded through the sale of services to install and support the system. Other models of funding for open developments were discussed at the conference, and Stephen Downes has outlined a few of these including:

  • Endowment model; where a project is sustained on the interest from an endowment.
  • Membership model; where a group of organisations make a certain contribution. The Sakai Educational Partners Program uses this model.
  • Donations model; where a project is sustained through donations. Wikipedia and the Apache Foundation use this model.
  • Conversion model; or service model, where the software is available for free and subscribers pay for support services. LAMS International uses this model.
  • Sponsorship model; where a commercial sponsor, institution or government sponsors a particular development. JISC funded projects fall into this category [4].

From a user and institution point of view one of the problems is that users often regard open source projects as “free”, particularly if they start life being funded by government institutions such as JISC. Glyn Moody pointed out that some of the confusion surrounds the word “free”, users might be thinking of the meaning: “free as in beer”, when the meaning “free as in freedom” is more relevant [8].

Jim Farmer from the Sakai Educational Partners Program said that open source projects needed to be much more aware of the needs of the institutions they were marketing their products at. And that the success of any open source product would depend on changing the behaviour of their customers.

In terms of developing user communities, Jim referred to the “Gladwell effect” in which Malcolm Gladwell identified the key elements of a thriving user community:

  • Connectors – influential people in touch with many groups
  • Mavens – experts in one area
  • Salesmen – to win new customers

It is also important that communities are able to harness people who start as “interested others” into playing a more active role. [9]

JISC and sustainability

Bill Olivier, Head of Technical Development for JISC is developing a sustainability white paper. At the conference he discussed the tension between the way JISC is funded through top-slicing and any sort of foundation or subscription mechanism. Would some institutions feel that they had paid twice for projects? Conversely, what about institutions who had paid for projects through top-slicing and then not used them at all? [10]

Ideas for how JISC could support project sustainability were suggested in a discussion session. John Norman from the University of Cambridge wondered whether JISC was in a catch 22 situation where top sliced funding was being used to fund projects that were too risky to be funded by individual institutions yet this means that projects would ultimately struggle to be sustainable once funding stops.

Earlier in the conference, John had suggested that the grant funding model does not help open source projects become sustainable, because it encourages competition rather than co-operation. Rather, he suggested that JISC and other funding agencies could fund people to contribute through three year fellowships, that would be peer reviewed, this would create an atmosphere of continuity and co-operation [11].

Sarah Porter, Head of Development at JISC stressed the need for Darwinian mechanisms. It is important that there is “survival of the fittest” amongst projects because with a limited budget it is not possible for JISC to continue to sustain existing projects as well as funding new innovations. She emphasized that projects needed to be responsible for developing their own user communities, and pointed out that there are many examples of software developments that aren’t even used in their host institutions. It is also the case that there are complex relationships between development projects and the final ‘marketed’ products, with many years between the two [12].

Tish Roberts describes how JISC will identify whether a project has developed that community:

“We have also been working on an open source maturity model (OSMM) which identifies those features (i.e. active up-to-date website) of an open source project that indicate that it has reached a threshold of usability by a community outside the one in which the project was originally developed."

While the Open Source Maturity Model is relevant for many JISC development projects, the difficulty for the toolkit projects is that they only last six months, too short a time to deliver the toolkit as well as build a user community.

Lessons from RELOAD

Perhaps toolkit projects could learn lessons from the RELOAD project. RELOAD is “a set of tools for editing IMS and SCORM compliant learning objects, metadata and learning designs” [13]. At the conference, Tom Franklin, from Franklin Consulting, outlined how the project team have considered various models for sustainability and have analysed their suitability for the project [14]. The project has decided to go down the sponsorship route, which fits the level of funding required and the user profile of the product. RELOAD has developed an international reputation and is regarded by some users as the reference implementation of the IMS specifications. It is also an integral part of commercial products, such as HarvestRoad [15] and projects like the European Ten Competence project [16]. Sponsorship is being sought from a number of these groups which include commercial companies and government agencies. Tom’s two messages for open source projects are:

“Firstly ask yourselves whether the project should continue at all – what need is it fulfilling? If you do decide to sustain the project – plan early, it will take a lot longer than you think to design and implement a sustainable model.”

From early adopter to mainstream

So if user communities are key to the success of open source initiative you might ask where is the user community for the e-framework and toolkits? Fortunately for the e-learning community several active developer and user communities already exist in the shape of the CETIS Special Interest Groups (SIGs). These are groups of early adopters largely from educational institutions with an interest in standards and interoperability. Many of the SIG members have been involved in developing toolkit and reference model projects and are therefore early adopters of the e-Framework approach and web service technologies [17]. JISC is starting to consider whether the SIGs should play a more formal role in the development of reference models and web service toolkits in each domain area. It will be important to co-ordinate the work of these groups to ensure that they are adopting similar solutions to problems which will mean interoperability will be realised in institutions.

There are already examples of early adopter institutions where IT staff are adopting a service oriented approach including the Enterprise II project at London Metropolitan University [17]. However experience tells us that widespread technology adoption is a tricky business. Many institutions will not have the capacity to implement these approaches without help.

One model would be for large vendors of MIS and VLE systems to adopt the service oriented approach, allowing institutions to gradually add functionality to their existing systems. JISC is starting to work with large vendors to elicit their thoughts on the framework developments and is planning an event with vendors in June 2006.

Another model is that small businesses that might develop to offer support services to institutions to help them deploy web service based systems. The idea is that small companies might emerge that offer distributions of a particular set of services and toolkits in a similar way to what happens with Linux. Scott Wilson again:

“One way the toolkits might get used by universities and FE colleges is through distributions. That is how Linux works, Linux is a collection of thousands of tools and integration kits, user interfaces, and there are organisations that basically pick a selection of those tools and assemble them into a deployment and distribute it on CD for instalment e.g Red Hat is a selection of Linux tools, it’s known as the Red Hat distribution. It’s a niche that companies could move into, but it wouldn’t be something that JISC would do because it gets into the realm of customer support, user manuals and doing user interfaces. In other examples distributions get started through a separate foundation, but it is an essential part of getting the technology used and exploited.”

Conclusion

Sustainability is clearly a key issue for the open source toolkit projects in the e-Learning Programme. In a narrow sense sustainability relates to the survival of ideas and continued development of code. But sustainability of the e-Framework initiative will rely on developing an active user community, toolkit projects are too small to do this themselves and this is where the CETIS SIGs and Reference Model projects are playing a key role.

In the long term however institutions will need help to combine services across domains to address their own business needs. JISC is starting to consider how it might best sustain these developments to ensure all institutions can benefit from service oriented approaches and e-Framework Initiative.

References

[1] e-Framework for Education and Research

[2] Sarah Holyfield Non technical guide to technical frameworks - part 2

[3] Sarah Holyfield Reference Models, the next important step

[4] Stephen Downes Models for Sustainable Open Educational Resources.

[5] Moodle on Wikipedia

[6] OSS Watch Open Source and Sustainability 2006 conference Programme and presentations

[7] James Dalziel's presentation at Open Source and Sustainability 2006, Why sustainability matters for open and closed source software

[8] Glyn Moody's presentation at Open Source and Sustainability 2006 Beyond Open Source

[9] Justin Tilton's and Jim Farmer's (instructional media + magic, inc)presentation at Open Source and Sustainability 2006: The Commercialization of Open Source Software.

[10] Bill Olivier's presentation at Open Source and Sustainability 2006 “Top Slice and Open Source”

[11] John Norman's presentation at Open Source and Sustainability 2006 Welcome

[12] Sarah Porter's presentation at Open Source and Sustainability 2006 Where Next?

[13] RELOAD project web site.

[14] Tom Franklin's (tom@franklin-consulting.co.uk), and Oleg Liber's presentation at Open Source and Sustainability 2006 Development of a sustainable model for the support of an open source toolkit

[15] HarvestRoad

..[16] TenCompetence project

[17] Christina Smart Large Scale deployment of web services at London Met

 

Supported by JISC Supported by CETIS
Powered by Plone