Semantic web session at CETIS

November 21, 2007

http://wiki.cetis.ac.uk/Semantic_Structures_for_Teaching_and_Learning

David Millard has described the semantic approach to sharing eLearning. His talk http://hoos-foos.blogspot.com/2007/11/why-semantic-web-hasnt-failed-and-how.html include the usual set of acronyms RDF, OWL etc. His analogy is with the Wizard of Oz and ends with a statement that we need to work together to build the yellow brick road brick by brick. This sounds like it might be a lot of work! The key being the adoption of RDF. The questions reflect my own doubts about whether this is actually the way forward when the web is a rather less organised. I felt that the very goos talk that Erik Duval gave at the OpenLearn conference showed how he thought the metdata world was moving on to an automated and useful tool. He stated that automatic metadata using context gave at least as good a result as manually produced. Perhaps this could apply to the generation of RDF (or whatever is looked for) from OpenLearn content, I know that Stefan Dietze of the LUISA project was considering this at the beginning of the year – I wonder if that has progressed (I will try to find out!)

 

Mikeal Nilsson http://mikaelnilsson.blogspot.com/2007/11/position-paper-on-semantic-technologies.html introduced a few more acronyms – SPARQL, GRDDL – these are trying to make things easier but do run the risk of confusing me more. His lead ppoint though was that the semantic web is just well designed web presence where you can “follow your nose”. He was asked if this was a push to a backstage approach rather than concentrating. This prompted me to mention Erik’s work.

 

Alistair Miles

Added a new acronym AKOS w3.org/2004/02/skos (and a variant acronym RDFa) but led with the tragedy of the commons – such as a shared fishing ground that nobody takes collective responsibility for. Alistar is from the eScience area and has spent 3 years working on SKOS – Simple Knowledge Organisation System. He suggested that skos could help eLearning as it would allow concept structure to be represented. He also covered rdfa.info as a way to put arbitrary RDF data into XHTML to reveal the structured data implicit in the content. is examole is links between people, places and events by pasing data from the web into a desktop application. (Possible link to Paul Mulholland’s work on StoryFountain and also to Grainne Conole’s looking at representing LD as concept web.)

 

How does web2.0 relate to semantic web? SKOS could be the bridge between the informal world of tags and the formal ontology based work.

 

Why has uptake been slow? Only just now beginning to understand how to apply this he drew an anology with programming but I am not sure it works as I feel that programming has become a niche and specialist activity and we need wider take up than that.

 

Michael Gardner firm Essex talked about applying a semantic approach to eLearning. He explained that he was not going to get too bound up in the definitions. His three examples are:

DELTA based on the earlier work for JISC on practitioner support. They built an ontology for subject areas by importing from Wikipedia’s OAI interface.

eProfile – a tool for visualing connections using foaf data, and has been developed into

ResourceBrowser – which was a hybrid between information about people in a social network and then connect to DELTA organised resources.

His third example uses a combination of natural language and document structure to find personal information – and then demonstrate this through 3d interface built with Sun’s darkstar/wonderland. The e.g. was that a record collection could appear as 3d cloud.

Lorna made the comment about how interesting it would be to see a concept map of all the documents of her machine. The clouds of words produced by perhaps show an indication of this. I will add one as a picture of this session!

 

Simon Buckingham Shum

Explained how his work built on how to give compelling learning experience to users and if semantic approach helps that happen then good. Simon’s approach is to focus on interpretations as a layer above the semantic web. His work of course has a deep basis and he showed Compendium (availabe via OpenLearn). Simon’s make great personal use of compendium and so was able to show his map of the discussion so far. He also showed Grainne’s LD compendium mao. A bit that I had not seen was his memetic work on modelling meetings in the access grid.

Simon also showed the soon to launch cohere based web tool. This could offer an impressive interface for building conceptual links.


Tagcrowd view of presentation

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: