Carsten Ullrich from Shanghai Jiao Tong University. This is a slightly adapted version of a talk he gave at the recent WWW2008 in China.

Carsten’s slides are available at http://www.slideshare.net/ullrich

Carsten is based in the e-learning lab – looking at how learning can be made easier and more interactive using technology. Recently they have been looking at Web 2.0 technologies/approaches in learning. What they found is that these approaches were transformational – you have to change the way you teach to use these approaches.

Carsten is going to cover:

• Motivation
• Web 2.0 from a learning perspective
• Web 2.0 as a research tool
• Examples
  • Social bookmarking for learning object annotation
  • Microblogging for language learning
  • Totuba Toolkit
• Lessons Learned

Starting with an outline of ‘Learning Management Systems’ – these support teacher centered ‘administered learning’ – if a lesson is ‘mastered’ then the student is allowed to continue – very typical ‘knowledge transfer’ paradigm, where teacher imparts wisdom to student.

An alternative approach is ‘Cognitive Tutors’ – built to support cognitive learning theories. These are very expensive to build – getting and encoding the principals of the theories into software is difficult.

What about Web 2.0? Often associated with constructivism – learner centred, emphasises collaboration and in-context learning, teachers provide assistance, advice etc.

When the e-learning lab looked at this they could not find an analysis of the technological foundations of Web 2.0 from an educational perspective. So first question ‘what is Web 2.0’ – looking at comments from Tim Berners-Lee and Tim O’Reilly – they adapted the characterisation of Tim O’Reilly.

Web 2.0 stimulates individual creativity – enabling and facilitating active participation. The value of a web 2.0 service increases the more people are using it.

From an education  perspective there is a large potential peer network.

The web provides diverse data on an ‘epic scale’ – huge variety of data, via browser and APIs, often annotated, and increasingly semantic and/or linked. From education – lots of information sources, access to data from real contexts which can be integrated into learning.

Web 2.0 supports the ‘architecture of assembly’. You can get students to combine data sources, but more importantly researchers/tutors can build rapid prototypes to try out.

Carsten illustrating what an iGoogle based ‘PLE’ (Personal Learning Environment) for language learning could look like – emphasising ease of ‘drag and drop’ approach iGoogle supports.

Web 2.0 takes a ‘perpetual beta’ approach – continual improvements/refinements to software. For education this can be confusing and distracting.

Some additional principles of web 2.0:

• Independent access to data
• Leveraging the Long Tail
• Lightweight models

Moving on, what can Web 2.0 do for research?

Lots of available services available (often free), which can easily be combined to deliver new functionality. Again, prototypes can be built very quickly.

Two examples:

Social Bookmarking for Learning Object Annotation

Authoring learning resources is time consuming and difficult task. Need to:

• Support lecturers with no or little knowledge about learning resources and metadata standards
• integration in existing workflow and LMS

So, they designed a method for lecturers to use delicious to bookmark resources, using predefined tags which represented:

• Concepts and relationships of subject domain
• instructional type
• Difficulty level

Used prefixes to allow effective filtering: “sjtu:”

Within the LMS for each page about concept c, look up resources in del.icio.us and add them on the page. Very simple – almost no effort to develop prototype.

The lecturer feedback was good. However, they found that students don’t look up external resources if the textbook is good enough – so need to concentrate on courses where the textbook doesn’t give enough coverage.

Micro-blogging for Language Learning

• Context: distant campus of SJTU – vocational learners: limited time, seldom active, shy
• Goal: provide practice possibilities
• Hypothesis: Micro-blogging
  • increases sense of community; reduces transactional distance to teacher (i.e. teacher is just another ‘peer’); can be done in very small amounts of time

The e-learning lab implemented a twitter-update downloader to store all twitter updates in database, plus automatic grading based on number of updates. This was based on the Twitter API – but there were limitations (could only extract the last 20 messages), so they had to do work screen scraping the web interface as well.

They got 98 students out of 110 participating; 5574 updates during 7 weeks. about 50% students sent 1-19 updates, and 50% sent 20-99 updates, but a few ‘power users’ who sent 100+

Only 5% of students felt it didn’t achieve the aims (sense of community etc.). 50% stated they communicated with native speakers – although Carsten thinks that only 5% were actually having real ‘dialogue’ in this context (based on log analysis).

The main criticism was that there wasn’t enough ‘correction’ of mistakes (interesting, because this seems to suggest students would have liked some more ‘knowledge transfer’ elements?

Lessons learned

Web 2.0 services can stimulate active participation – Twitter usage continued after lecture. But saw users drifting ‘off topic’ – e.g. posts in other languages, pejorative messages

They found that students didn’t make use of the ‘architecture of assembly’ – e.g. didn’t reuse to show on blog.

Totuba Toolkit

Totuba is a startup providing a toolkit for schools, university, researchers.

What it does is provide a ‘Research Assistant’  to enable capturing, categorising and referencing of information; provides social network tools; share information; store; export to word etc.

Carsten is showing how this might work with a Wikipedia article.

He is comparing to Zotero – but the idea of Totuba is that it is very simple – it can be used by school children, you don’t need to install a plugin etc.

Looks like this is in alpha at the moment

The goal is to facilitate the process of learning and research, removing unnecessary steps, automating manual integration work, and make it easier to find additional materials and peers.

The concern I have with this is whether it is reinventing existing services, or whether it adds some value to yet available? So, what would make me use/recommend Totuba over Google Notebook + Facebook + … I guess simplicity and packaged product is the answer, but this seems to conflict with the rest of the message from Carsten. Need to reflect on this more.

Conclusions

Web 2.0 tools and learning:

• Less suited for designed instructions
• have potential to stimulate active participation
• learniers will think of anticipated ways of usage
• requireds active teacher input/monitoring
• you can become dependent on 3rd party applications

Web 2.0 and research

• functionality at high level of abstraction
• quick way to assemble prototypes
• one becomes dependent of third party tools
• difficult to find data about scientific publications
  • Google scholar: no API
  • Citeseer, DBLP: restricted to Computer Science
• Open Linked Data: still for experts

Question: Do these compliment or replace traditional methods

Answer: These are transformative – so eventually supplant traditional methods

Question: How to you do assessment? (cynically
you can say that students are motivated to ‘get the qualification’ not ‘to learn’)

Answer: Still had conventional exams in these examples. Twitter usage was measured, and contributed to mark – to get participation

Question: Are you saying this approach is more inline with human nature?

Answer: Yes! That’s a good summary.