Thursday, May 31, 2012

Dev8ed (day 2)

Dev8ed (day 2)

After the presenter pitches at the beginning of day two of the first session left me in a bit of a quandary - there were two presentations that I really wanted to see at the same time. One was from Adam Hyde, the project lead for BookType, which is a piece of software which is used by the free/libre/open source software (FLOSS) community to collaboratively create software manuals. BookType could be thought of as a mash up between Google Docs and Wikipedia - basically it is a tool for distributed groups of people to collaborate online to create conventional books. Books created using BookType can be published in all the standard electronic publishing formats or exported directly to Lulu for publication as a print on demand book. The world publishing is rapidly changing as more and more information is made available under free to reuse licences such as Creative Commons, and the future of university textbooks may well be with tools such as BookType that allow a group of authors to collaborate on rapidly producing books. Although Adams talk title of "How to write a textbook collaboratively in five days" was intriguing I decided that I would learn more by attending Diana Laurillard's session on learning design.

Diana, along with Patricia Charlton, was presenting a tool called Pedagogical Pattern Collector which is used to collect and abstract learning designs which can then be adapted and reused. This is nothing to do with the IMS learning design specification, but a much more teacher orientated concept of a learning design. Software developers will mostly be very familiar with the concept of design patterns, established solutions to frequent types of problem which may appear different due to their context but have structurally similar solutions which are best achieved by following the established patterns. Diana is attempting to achieve something similar with teaching practice, and provide a way of scaffolding teachers planning of lessons and courses, and helping teachers make creative and effective use of technology enhanced learning. The software runs online, and has two modes: a browser of which lets you search for and look through existing learning designs, and a designer that let you adapt designs or put in new ones from scratch. I felt that the software could be excellent as a way of helping training of University teachers and GTAs, and also for student teachers, and Patricia told me that they have already used the software for this purpose. Unfortunately the URL of the online demo we used looks very temporary, so I will not link to it here. There is a different (older?) version available here. I'm sure that Diana and Patricia would be delighted to show the current version off to any academics in the UK who wish to play with it.

The next session I attempted to attend was "Learning Technologies and Historical Perspective", but the presenter didn't show up so it ended up with me, Simon Booth and Guy Pursley sitting discussing mathematics education, and the way that by practising solving mathematical problems the brain seems to develop an ability to recognise patterns, and therefore apply the correct solution. There is a lot of similarity between mathematics and software development, and maybe also with teaching.

The final session I attended was on XCRI-CAP, a specification for sharing course catalogue information online in a machine readable format. XCRI-CAP is a bit like Atom or RSS in that it allows a college or university to publish their course information in a form which can be used by an aggregator to make larger course directories, very much like the way that Atom and RSS allow aggregators to pull together blog feeds. XCRI-CAP is backed by JISC and is likely to become a British standard very soon, so it seems likely that if the not very distant future it will become essential for Universities to provide XCRI-CAP descriptions of courses to help with recruitment of students. Adult and continuing education departments are likely to need to make use of XCRI-CAP aggregators to get access to students who are using the public interfaces of these tools to search for courses by location rather than by institution.

Dev8ed (Day 1)

Dev8ed, a two day meeting for software developers working in H&FE, took place at Aston University on the 29th and 30th May 2012. Like the HEA conference there was so much of interest that it was difficult to choose which sessions to attend, but the organizers did alleviate this by having presenters doing pitches where most conferences would have had a plenary, and many of the presenters also did 'lightening talks' - 10 minute mini presentations on their work.
The first session I attended had presentations Fridolin Wild from the Open University and Martin Hawksey from JISC CETIS. Fridolin presented on Widget based personal learning environments which I suspect will be something of increasing importance over the next few years, and may lead to the merging of VLE, e-Portfolio and student portal applications into a single personalised site. Martin has been using Google Spreadsheet as a powerful data acquisition and presentation tool - his blog is well worth a visit to see some of the things he's done. Because Google Spreadsheets can acquire data from twitter this could be a powerful tool for looking at social networking behaviour.
During lunch I had a chat with Chuck Severence who moved from IMS to Blackboard about the same time that Blackboard bought moodlerooms. His views on Blackboard's attitude to Moodle and Sakai are reassuring - their interest in open-source relates to expanding service provision, not getting more customers for their VLE, and their new subsidiary companies will continue to contribute actively to OSS.
In the afternoon I attended a session on IMS Learning Tools Interoperability (LTI) by Simon Booth and Stephen Vickers from the JISC ceLTIc project - they have added an LTI link to WebPA (a group work peer-review tool) and will be offering it as a hosted service that other Universities can link to using LTI links from their VLE. To me the great thing about LTI is that it provides a safe way of extending the functionality of the VLE without any risk to the core service - this give great potential for experimental learning technology projects. Stephen has written PHP classes to support LTI and during the course of the meeting Steve Lay almost completed a Python implementation. There were some questions about the possibility of a Perl implementation... Over the next few weeks I'll be liaising with Stephen and Simon to make sure my Java and C# implementations have as similar an API as is practically possible to Stephen's classes which will hopefully allow a single documentation set. Once we have solid and easy to reuse, liberally licenced PHP, Java, C# and Python implementations available I think we should satisfy the needs of most normal developers. I expect someone will create a Perl implementation for the abnormal ones.
I'd have liked to see the session Chuck presented (in his University of Michigan role) on Coursera a massive open online course (MOOC) system. This is interactive distance learning on a grand scale - Chuck's Internet History, Technology, and Security class has approaching 2000 students. Unfortunately the LTI discussions went to too long, so I joined the end of a lightening talks session instead. Malte Ressin's talk on internationalisation reminded us that there's more to internationalisation of software than language - cultural differences mean that colour choice can have quite different impacts, and images that are perfectly acceptable for us can be taboo in other cultures.
Steve Lay talked about OData, a web service data protocol that sits somewhere between the complexity of SOAP and the relative chaos of REST. The OData approach seems to be that they take the best ideas and practice from existing specifications that actually work in practice and standardise them.
The next talk was mine - I briefly described my Java and C# implementations of LTI and demonstrated setting up a quiz running on the Amazon EC2 cloud from the Services Moodle.
I was followed by Guy Pursley from Reading who has been doing interesting things with the Blackboard e-portfolio tool. They wanted a more guided approach to creating a view than Blackboard provides, so Guy has written JavaScript code that guides the student through the process and provides additional hints. Because the code is all client side it should be possible to adapt it to work with Mahara as well. Guy is also keen to get Leap2a support into the Blackboard tool, so hopefully my Leap2a validator will be of use to him.
Alex Iacconi discussed cloud platforms - particularly the ones like Google App Engine and Windows Azure - there's something very appealing about a hosting system that scales as required.
Diana Laurillard gave a brief talk on Learning Designer - more on that later when I blog about day 2 of the meeting.