Jennifer Zaino
SemanticWeb.com Contributor

What’s happening inside the Semantic Technologies for the Enhancement of Case-Based Learning, the big U.K. project to enhance 21st century learning we reported on earlier this summer? To find out, Semanticweb.com recently caught up with Dr. Patrick Carmichael, project director and head of the Evaluation Group at CARET, the Centre for Applied Research in Educational Technologies at Cambridge University.

Semanticweb.com: How did CARET come to be involved in this project?

Carmichael: CARET’s mission is to support teaching, learning, and researching across the university, but also to take on broader projects like this — externally funded research. Historically we have tended to do broadly educational technology projects concerned with developing virtual learning environments. But I also head up the unit concerned with developing teaching and learning with and without technology, and some of the projects have very minimal tech components — developing approaches to teaching more generally, for example.

So, one of the interesting things is that we have people who have an educational and social sciences background, and people who are software developers and content developers under the same roof. This new project is interesting because it really is specifically designed to be interdisciplinary, where the technologists and social scientists are working and learning with and from each other. That is central to the project.

And we’ve had a number of projects in which you can probably see the genesis of this one, such as digital repositories that have supported a lot of the big digitization projects at Cambridge in the university library and museums, for long-term digital preservation. One of the key applications we’ve identified there is the Fedora digital repository, which is very well set up to be adaptable for semantic web applications.

Also, we’ve had a number of teaching and learning projects to do with teaching complex or rapidly-changing or controversial issues — we have teaching staff very interested in how you prepare undergraduate and post-graduate students to deal with complexity through problem and case-based learning. That’s where the project emerged for us, from a dual interest in developing robust and scalable architectures and doing really high level and effective teaching around complex issues. The semantic web fits in as way of supporting both our technology concerns and those teaching and learning issues.

Semanticweb.com: What are the specific roles for your group?

Carmichael: There are two specific roles for our group. One is that this is where the actual technology development is going to take place. We will be drawing on our prior experience with digital repositories, linking with the MIT team [that developed SIMILE (Semantic Interoperability of Metadata and Information In unLike Environments)], and also we have a very close working relationship with the Economic and Social Data Service, which is the electronic digital repository for social sciences in the U.K. They are the long-term guardians and brokers of social science data from research projects in the UK. So Cambridge is at the center of the group of technology developers and service providers, within the project

Also we have three or four departments at Cambridge signed up for the project. They will help us develop domain-specific applications of semantic web technology we’re interested in. So the other job is working intensively with teachers and students to actually establish what particular aspects of the semantic web have potential to enhance teaching and learning — that is, to distinguish between the theoretical vision of the semantic web and the practical uses of semantic web technologies that teachers and students latch on to. Maybe they are not very concerned with the broader vision of the semantic web but they are interested in what it lets them do to support their teaching and learning. One of the critical jobs is to work with groups of teachers and students to establish what they think the suite of semantic web technologies will actually offer them. That goes beyond improved resource discovery: it’s about being able to collaborate around complex issues and take part in knowledge construction activities.

Semanticweb.com: Can you give us an example of how this is playing out?

Carmichael: We have some pilot projects at the moment. For instance, we are bringing in undergraduate researchers to work in the summer, and have four working at the moment, doing development work in their own departments. They’ve been taking some elements of the SIMILE toolkit and are liaising with teachers in their departments to try and develop proof of concept applications with potential in teaching and learning. They have identified some interesting niche applications and are building prototype web applications. That’s been very successful in engaging and testing out what is required.

Take, for example, the group of students and staff in bio-sciences. They are looking at the spread of plant diseases.. There’s quite a well documented problem in California with a disease called Sudden Oak Death, and they have been able to gather together lots of partial data sets and information from other university projects that would let undergraduates engage with all these different data sets and build theories up about what is contributing to the spread of the disease — is it about the particular species, weather conditions, soil types and so on They convert the data sets to RDF, use SIMILE to aggregate them, and the SIMILE timeline and Google Maps API to display information and will allow students to formulate hypotheses.

We also have a document base of 17th century documents about the area around Cambridge and events during the English Civil War. That’s a nice semantic web demo because it lets you pull together biographies, historical, and geo-graphical information and look at these documents in those contexts. You can look for new patterns using a combination of semantic web and data mining, and we are actually generating some tag clouds around individual documents to do document similarity measures as well.

There’s a real buzz around these projects, and as the staff becomes skilled up and start thinking about semantic technologies, we’ve seen transfer of ideas and approaches around different projects: concepts, technologies and solutions all get swapped around.

Semanticweb.com: What has surprised you so far?

Carmichael: One of the things we found very important, and will have to build into the big project when it starts in October, is how important it is to really keep open channels for communication.

We originally thought we’d have some kind of user requirements process that would act as a conduit between domain specialists and technologists, and we’re actually finding they are forming into little groups themselves. Our technologists and programmers are spending time in different departments. One of my researchers employed in the undergrad program — I expected him to work next to me and he’s actually across Cambridge working in the Department of Plant Sciences, because he feels he needs to be working with teachers and responding to fast-developing user requirements.

The more our teachers and students learn about the technologies, the more they realize they can do, so the user requirements from three weeks ago are completely outdated. There’s a rapid learning curve. As a team we get together every week and we tend to find the distinction between the developers and programmer and domain specialists is blurring and it is sometimes quite difficult to establish who is who!
Personally, I haven’t seen this kind of buzz since the early days of the web, when people discovered they could develop web pages and there was an explosion in user generated content and sharing of technologies, solutions, and applications. Suddenly people in education are realizing something really new is available to them.

This comes back to what CARET has been doing for a while. We’ve always been cautious about being too technology-led and then our work is just about embedding the technologies into the institutional context, because then it can easily slip into coercing people to use technology they are not keen on. So there’s an ongoing process of dialogue, where as we explore, we are finding out an awful lot about what historians and plant scientists and others do by their taking part in development processes. Alongside developers. If we can maintain this kind of impetus in the three-year project and generate semantic technology enabled applications which have an impact on teaching and learning as a result, I will be delighted!

SemTechBiz is Less Than 2 Weeks Away

The Semantic Tech & Business Conference (SemTechBiz) is coming to San Francisco on June 3-7! Join us for case studies, innovative panels, tutorials, and keynotes that will provide you with practical advice, hands-on guidance, and breakthrough approaches to solving business problems with semantic technology. Passes go up $200 at the door. Sign up now and save !