JISC CETIS Conference 2007 (2)
The 2007 JISC CETIS conference was held in Birmingham this week, spanning a two-day event comprising of three keynotes and ten workshop sessions.
I went to the Semantic Structures for Teaching and Learning session on the second day.
Semantic Structures for Teaching and Learning
The semantic session started with short presentations from six speakers, with each providing an insight into different semantic areas. Tore Hoel started with a look at moving Towards a Semantic Infrastructure for Learning, Education and Training?, considering that current slow progress could be remedied by factors such as more trust and reusing vocabularies more often. Tore also explained the use of a ‘backstage’ for Norwegian websites, where sites are encouraged to expose technology, ontologies, etc, and gave an example of Kuling.net where users can expand existing ontologies. In conclusion, Tore proposed the need for simple recommendations to facilitate the exchange of ontologies and vocabularies, creating a semantics hub for identifiers, and demonstrating tools and making them known. View Tore’s presentation on SlideShare.
David Millard talked next about Why the Semantic Web hasn’t failed, and how we shouldn’t fix it. In his presentation, David advocated the promotion of RDF REST interfaces by ‘semantic people’, adding that they could also use popular ontologies such as FOAF and SKOS, and simply be a semantic extrovert. E-learning people could map ontologies automatically and capture folksonomies to aid semantic development. View David’s presentation on SlideShare.
Mikael Nilsson followed with a presentation on Standardizing Semantic Technologies for Learning. Focusing on areas such as web architecture and stability architecture, Mikael talked about novel methods of getting at data, mentioning the use of GRDDL. Mikael also advised that part of the semantic web is just well-designed ‘web presence’ – a “follow your nose” approach.
Following Mikael was Alistair Miles, who spoke about SKOS and RDFa in eLearning. Covering SKOS and RDFa, Alistair informed the audience about both technologies, providing an insight into both languages. He described how conceptual structures could be expressed directly in SKSO and used in Semantic Web applications, and advised that RDFa could be used to reveal structured data implicit in content, which could be more interesting that simple rendering of pages.
The penultimate speaker was Michael Gardner, who gave a presentation on Using Ontologies in eLearning. Michael spoke briefly about three instances – including Delta and ResourceBrowser – and explained a little about the projects. Michael concluded that ontology issues included ownership and quality control, the role of folksonomies and exploring new ways to build classifications.
Simon Buckingham Shum was the final presenter, giving a talk on Hypermedia Discourse: Semantic Structures for Sensemaking. During his talk, Simon promoted two semantic tools, Compendium and Cohere, highlighting the pros and cons of trade-offs with Compendium and describing Cohere as a “web tool for connecting ideas”. More information on these tools is available at the respective websites.
After the presentations (and a much-needed coffee break), the session returned to a discussion session involving the speakers as panelists. A lot of interesting points were raised – too many to be covered in full depth here – and prompted interesting debates among the panelists and audience. Some points:
- Semantic web is about allowing machines to exchange information with each other
- Not one semantic web but many with local coherence
- User interfaces should gather data at users’ level of familarisation
- Not enough semantic data available to build tools
- Viral grass roots spread – instant gratification – more fun experience=>”I’ll do it”
The quickfire presentations were an informative introduction into a topic that is worth looking into in more detail and the resulting discussion generated interesting points worth exploring further.
Entry filed under: cetis-2007-conference-semantics. Tags: .