After two years... preparing Phd Plan! (1)

Yes... I know that it can sound a bit weird... It's something that I should have done during the first year of my PhD... but there is always a positive side... now I have deeper knowledge in my field. So now, I have to do a bit of inverse engineering for what I have got so far and a bit of planification for my next challenges... The goal of this reflection is finding the research question that I have answered and the research questions that I am going to answer in the near future.

So let's go to think about what I have so far...

At lirias you can find my publications at KULeuven.

In 2010, I got two publications. The first one was presented at ECTEL conference. It describes the ARIADNE architecture. My contribution to this paper is about the ARIADNE registry and it has a strong link with the second paper due to it is the description of the already mentioned ARIADNE registry that we presented at the SE@M workshop.

What is this registry about?

The ARIADNE Registry is an open source project that enables users and services to discover new repositories of learning resources. Yes... I know... it sounds interesting... isn't it? ;)

What can we do once that we have found a new repository?

First, I should point out that these repositories can work in a federated way... why? Because they can share descriptive information about the learning resources and, to this end, they expose this information through service interfaces. This idea is similar to the WEB 2.0 concept, different services that can interact with others, for instance, facebook has an application that if you install it, all your tweets are posted directly in your facebook wall. It's cool... updating one social network status, it appears in two social networks... so one time work... double impact of your status...

Let's apply the same concept to the repositories... you (as learning resource provider, simple user or teacher) update your learning object once and such information is extended along the network of repositories hence your information is accessible from different access points. (It's important that user remains the control of the object itself... so this federation is not incompatible with business models). The idea follows the same concept: increase the impact of your work!

It seems ideal the situation but...

What is the challenge?

The challenge is that repositories expose their information using different protocols and specifications. For instance, repositories can expose their information using SQI, SRU/W or OAI-PMH protocols and the description or metadata using specifications such as IEEE LOM, Dublin Core and RSS. Therefore, these requirements should be described somewhere enabling access to external services to this information automatically. In this way. they can understand how they can access to such metadata.

Looking for a possible solution in the context of ASPECT project, we used the specification IMS LODE. Such specification allowed us to define different collection of objects and define the protocols to access the repository information.

It worked quite well providing access to more than 700.000 metadata objects in total and the integration fo the ARIADNE Harvester with the ARIADNE Registry.

What are the limitations of this approach?

Currently, learning resources are not static objects such as docs and pdfs. Blogs, tweets, forums and so on can be learning resources. IMS LODE is an specification oriented to define protocols and specifications, however these WEB 2.0 services do not follow specifications and protocols, so most of the definitions are specific to such services and we can lose the process automatization.

What is the research question?

What are the benefits on defining services to share learning resources using IMS LODE?

As you will see in my next posts, I changed a bit my orientation on my research focused on how we can track users and enabling reflection on the activity for teachers and students, but following the same research line 'mashup of services'.