Snapshots of standards and technologies in use in the DVLE programme

The JISC DVLE programme is well underway now, and as part of our support role for the programme we have been discussing with the projects, the different approaches, technologies and standards they are hoping to use. A record of the discussions is stored in our PROD project database. We find these discussions a really useful way for us to get a clear idea of what projects are actually doing – as opposed to what they say they are going to do in a project plan.

PROD is also growing into a substantial record of the technological approaches from a growing number of JISC programmes, almost a collective memory if you like. Over the past few months my colleague David Sherlock has been developing some easier ways to get information out of PROD and provide some visualisations of the data we are recording, you can read more in his blog. So, in relation to the DVLE programme here are a couple of snapshot views of the data we’ve been recording.

Firstly a wordle of the standards and technologies. Quite interesting surface overview, but doesn’t give much detailed information.

DVLE standards and technologies wordle
DVLE standards and technologies wordle

A mindmap showing each project entry, you can click through to moved down from project name to standards/technologies and then comments. I think this provides a useful, digestible summary of the programme. We’d like to develop this more to include links to project home pages, date stamps for comments etc.

And yes, we will be creating a wookie mindmap widget but it was just quicker to use this existing google one for proof of concept.

A manyeyes view of the numbers of each standard/technology. This allows us to show the numbers of projects using each standard/technology. I think this could be increasingly useful to use across programmes to allow us to start building richer pictures of emerging trends.

We will be developing more of these visualisations over the coming months so watch this space and you can of course view the complete entries directly in PROD.</p

Getting down to the business of building distributed virtual learning environments

Over the past few years we have been following and developing the notion of distributed learning environments. This culminated earlier this year with the publication of the CETIS briefing paper on Distributed Learning Environments and the JISC DVLE programme.

Yesterday all eight of the funded projects made their way (well, actually the level of rain made it feel a bit like swimming) to Bolton for their first meet-up. The programme is divided into two strands, with the first comprising of three projects of six months duration, finishing in December this year. Glasgow Metropolitan College and Glasgow University and concentrating development of a specific widget/VLE plug-in each. Teesside University is taking a more user centric approach by running a number of workshops and then developing widgets from ideas that emerge from them. The other strand is made up of the remaining five projects (MMU, University of Reading, the Open University, University of Edinburgh and Southampton University). These projects are funded for a year, and are investigating the larger issues of integrating more flexible and interoperable approaches to institutional learning environments. More information about the projects is available from the JISC website.

The main technologies in use across the programme are W3C widgets (mainly through using Apache Wookie incubating), Open Social and IMS LTI and Basic LTI and their new Basic Outcomes. To help bring everyone up to speed on each of these a large part of the morning was given our to presentations on each. Ross Gardler (OSS Watch) did the honours for Open Social, Scott Wilson (CETIS) for Wookie and Wilbert Kraan (CETIS) for IMS. Copies of the presentations will be available via the CETIS programme support page over the next day or so. These set the scene for a round of breakout discussions. Mark Stubbs has already blogged about some of the specific security/ authentication/wookie issues the W2C project is grabbling with.

I sat in on the Open Social group where the security issue also came up. Ross argued quite strongly that the technical issues around security have to a large extend been solved outside the education sector and we just need to trust the technology. However, the group did agree that there were cultural issues with education (particularly HE) around knowledge and understanding of identity and authentication which needed to be broken down. We also discussed the possibilities of using open social in a portfolio context. The University of Reading are looking to extend the functionality and interoperability of their in-house developed portfolio tool. The group discussing the IMS options spent quite a bit of time musing over the time/cost implications of developing full LTI integrations over using Basic LTI and the limitations of both – from the wider when will the spec be finalised issue to smaller I can build it but how long will it last, and in the long run does that actually matter?

In the afternoon we had more discussion particularly around wookie implementation. One concern around wookie for a number of projects was its sustainability. As with any (relatively) new technology, sustainability of external systems is a key concern for anyone looking to deploy it in a significant context. However, as Ross pointed out more than once, the fact that wookie in now in part of the Apache foundation, the chances of sustainability are greatly increased. The University of Bolton are also committed to its development and again as with anything, the more use it gets the stronger it becomes.

Along side the more technical discussion there was a concurrent discussion around user-engagement. As explained earlier the Teesside project is very much focussed on gathering real user needs and has designed face to face work shops (adapting templates creating by the RLO CETL and the Sharing the Load project). We discussed many approaches to “paper design” including having print outs of various mobile devices to remind people of the actually size of the finished app/widget. The group all agreed that scope creep, nicely illustrated by Scott Wilson from some recent workshop experience where they found delegates trying to design a whole VLE instead of a specific “thing”, was something that teams needed to be mindful of.

W2C is taking a different approach towards user engagement. They are using an external company to build their first official iphone app (due out sometime next week), followed later in the year by blackberry, android and widget versions. The team are going to use this initial app with students and staff to get feedback and inform future developments. The cost of external development they feel is offset by time savings for the team and gives them something tangible to test with. There also seemed to be a general consensus that actually seeing “the app for that” was incredibly powerful in terms of user engagement – particularly for VCs 🙂

There is a great deal of synergy between the projects and I hope that yesterday provided an opportunity to forge stronger relationships across the programme and beyond. It looks like there will be a number of apps/widgets to share with the community by the end of the year.

CETIS is providing support to the programme and we will be organising a number of open meetings over the next year for other to engage with the projects. So watch the space for more updates and information and if you are involved with similar work, please let us know.

cetisdle – presentations now online

Over the past year or so we’ve been doing a lot of thinking and work around what we’ve now come round to calling distributed learning environments. Essentially, ways that you can extend current VLE functionality without having to change/upgrade your VLE.

Concurrently it also seems that every HE institution in the country is either about to start, in the middle of, or has just completed a review of its learning environment provision. So despite many calls of its death, it does seem that the VLE is going to be around (in some shape or form) for a while yet.

Last week in Birmingham we held an event to launch our DLE briefing paper. The event also coincided with the Distributed Virtual Learning Environments call announcement from JISC. During the day we had presentations demoing a number of the models featured in the briefing paper as well as an overview from Peter Hartley of the ALT LER (learning environment review) SIG. Heather Williamson from JISC was also able to give an overview of the DVLE call.

Over all there was a lot of interest from participants in exploring further ways to easily extend functionality of learning environments for users (staff and students) – particularly the development of widgets. Over the next year we are planning to run a number of events where we can bring developers and users together, so watch this space, and if you have any suggestions please let me know.

Presentations from the day are now available on the CETIS wiki. You can also view the twitter stream from the day.

The headless VLE (and other approaches to composing learning environments)

CETIS conferences are always a great opportunity to get new perspectives and views around technology. This year it was Ross MacKenzie’s somewhat pithy, but actually pretty accurate “so what you’re really talking about is a headless VLE” during the Composing Your Learning Environment sessions that has resonated with me the most.

During the sessions we explored 5 models for creating a distributed learning environment. :
1 – system in the cloud, many outlets
2 – plug-in to VLEs
3 – many widgets from the web into one widget container
4 – many providers and many clients
5 – both a provider and a client
Unusually for a CETIS conference, the models were based on technologies and implementations that are available now. (A PDF containing diagrams for each of the systems is available for download here)

Warwick Bailey (Icodeon) started the presentations by giving a range of demo of the Icodeon Common Cartridge platform. Warwick showed us examples the plug-ins to existing VLEs model. Using content stored as IMS Common Cartridges and utilising IMS LTI and web services, Warwick illustrated a number of options for deploying content. By creating a unique url for each resource in the cartridge, it is possible to embed specific sections of content onto a range of platforms. So, although the content maybe stored in a VLE users can choose where they want to display the content – a blog, wiki, web-page, facebook, ebooks etc. Hence the headless VLE quote. Examples can been seen on the Icodeon blog. Although Warwick showed an example of an assessment resource (created using IMS QTI of course) they are still working on a way to feed user responses back to the main system. However he clearly showed how you can extend a learning environment through the use of plug-ins and how by identifying individual content resources you can allow for maximum flexibility in terms of deployment.

Chuck Severance then gave us an overview IMS Basic LTI and his vision for it (model 2). Describing Basic LTI as his “escape route” from existing LMSs. LTI allows an LMS to launch an external tool and securely provide user identity, course information, and role information to that tool. It uses a HTTP POST through the browser, secured by the OAuth security. This tied in nicely with Warwick’s earlier demo of exactly that. Chuck explained his visions of how LTI could provide the plumbing to allow new tools to be integrated into existing environments. As well as the Icodeon player, there is progress being made with a number of systems including Moodle, Sakai and Desire2Learn. Also highlighted was the Blackboard building block and powerlink from by Stephen Vickers (Edinburgh University).

Chuck hopes that by providing vendors with an easy to implement spec, we will be able to get to the stage where there are many more tools available for teachers and learning to allow them to be real innovative when creating their teaching and learning experiences.

Tony Toole then presented an alternative approach to building a learning (and/or teaching) environment using readily (and generally free or low cost) available web 2 tools (model 3). Tony has been exploring using tools such as Wetpaint, Ning, PBworks in creating aggregation sites with embed functionality. For example Tony showed us an art history course page he has been building with with Oxford University, that pulls in resources such as videos from museums, photos from flickr streams etc. Tony has also be investigating the use of conference tools such as Flash meeting. One of the strengths of this approach is that it takes a relatively short time to pull together resources (maybe a couple of hours). Of course a key draw back is that these tools aren’t integrated with existing institutional systems and more work on authorization integration is needed. However the ability to quickly show teachers and learners the potential for creating alternative ways to aggregate content in a single space is clearly evident, and imho, very appealing.

Our last presentation of day one came from Stuart Sim who showed us the plugjam system he has been developing (another version of model 1). Using a combination of open educational standards such as IMS LTI and CC, and open APIs, plugjam allows faculties to provide access to information in a variety of platforms. The key driver for developing this platform is to help ‘free’ data trapped in various places within an institution and make it available at the most useful point of delivery for staff and students.

So, after an overnight break involving uncooked roast potatoes (you probably had to be at the conference dinner to appreciate that:-) we stared the second half of our session with a presentation from Scott Wilson (CETIS and University of Bolton) on the development of the Wookie widget server and it’s integration into the Bolton Moodle installation (another version of model 1). More information about Wookie and its Apache Incubator status is available here. In contrast to a number of the approaches demoed in the previous session, Scott emphasised that they had chosen not to go down the LTI road as it wasn’t a generic enough specification. By choosing the W3C widget approach, they were able to build a service which provides much greater flexibility to build widgets which can be deployed in multiple platforms and utilise other developments such as the Bondi security framework .

Pat Parslow, University of Reading, then followed with a demo of Google Wave (model 4) and showed some of the experimental work he has been doing incorporating various bots and using it as a collaborative writing tool. Pat also shared some of his thoughts about how it could potentially be used to submit assignments through the use of private waves. However although there is potential he did emphasise that we need much more practice to effectively judge the affordances of using it in an educational setting. Although the freedom it gives is attractive in one sense, in an educational setting that freedom could be its undoing.

We then split into groups to discuss each merits of each of the models and do a ‘lite’ swot analysis of each of them. And the result? Well as ever no one model came out on top. Each one had various strengths and weaknesses and a model 6 taking the best bits of each one was proposed by one group. Interestingly, tho’ probably unsurprising, authentication was the most common risk. This did rise to an interesting discussion in my group about the fact that maybe we worry too much about authentication where and why we need it – but that’s a whole other blog post.

Another weakness was the lack of ability to sequence content to learners in spaces like blogs and wikis. Mind you, as a lot of content is fairly linear anyway that might not be too much of a problem for some:-) The view of students was also raised. Although we “in the know” in the learning technology community are chomping at the bit to destroy the last vestiges of the VLE as we know it, we have to remember that lots of students actually like them, don’t have iphones, don’t use RSS, don’t want to have their facebook space invaded by lecturers and value the fact that they can go to one place and find all the stuff related to their course.

We didn’t quite get round to model 5 but the new versions of Sakai and Blackboard seem to be heading in that direction. However, maybe for the rest of us, the next step will be to try being headless for a while.

Presentations and models from the session are available here.

Widget meetup, London 13 October

Just to give some notice to a widget meetup we’re organising in conjunction with the JISC repositories team in London on 13 October. It will be a one day meetup for people who are interested in creating Widgets, sharing ideas about Widgets, and turning applications into Widget containers.

Scott Wilson (CETIS) will be there to talk about the Apache Wookie (Incubating) widget engine and progress on the W3C’s Widgets specifications, and to help anyone looking at integrating Wookie with other applications. Wilbert Kraan (CETIS) will also be demonstrating integration of Wookie with Google Wave.

The event will be fairly informal with lots of opportunities to share code and experiences; there will also be the opportunity to demo work in progress. If you have any suggestions for a short presentation please let me know.

The event is free to attend, and more information including a link to register is available on the CETIS wiki

ICOPER Widget workshop

Earlier this week Scott Wilson and I attended an ICOPER widget workshop in Vienna. We were invited to the workshop to give an overview of the CETIS widget workgroup and in particular the work Scott has been doing on the Wookie widget server.

The purpose of the workshop was to clarify the concept of widgets in the ICOPER context, particularly how to link widgets to competencies. One of the main drivers of the project is to build best practice around the use of digital technologies, so it was useful for the project partners to have an opportunity to discuss just what their expectations/understandings of widgets are before deciding on a collective way forward.

The morning was taken up by a number of presentations giving an overview of some developments in this area and some potential areas for development. Scott started the day and emphasized the need for interoperability for widgets so that they can be deployed across multiple platforms and aren’t subject to vendor lock-in. This is one of the reasons the CETIS working group has initially focused on infrastructure issues as previously reported. The work he has been doing for the Ten Competence project has been focussing on utilizing widget technology to add extra capability (for both teachers and learners) quickly without having to modify an institutional VLE.

As the day progressed it become increasingly clear that interoperability is key for the successful development and use of widgets in an educational context. A number of presentations illustrated use of page aggregators such as netvibes. In one way these are great ways to quickly and easily start building personal spaces (or personal learning environments – but I use that term with caution). However the business model of any of these services could change overnight and how would that affect use – particularly in an educational context? Would you pay for your netvibes page? More importantly how/where would you recreate that page – could you export your “widgets” elsewhere? I’ve already been burnt by this kind of changing business model my love affair with the Sprout Builder widget application came to an abrupt end when they monetized their business model. I really liked their widget building wizard, but for my needs, it just isn’t really viable to pay their monthly subscription. Maybe one thing the ICOPER community could look at would be building a free, open-source widget editor.

One of the key things about widgets that seems to have a general agreement, is that they are small applications that generally do one thing, but do that one thing very well. I was slightly concerned when discussions came around to developing an e-portfolio widget. However I think what was actually being proposed was exploring how/if widget technology could provide different views into a portfolio, which actually might be useful. You may only want to show/share a small part of your portfolio any one set of users. I was more taken with the work being done by the OUNL with open content where they are developing widgets to aid reflective learning. They described there work as developing “learning dashboards” where information such as how long you had spent on a course compared with the average user study time is available. This information is easily available in most VLEs but very rarely shared with students – only a tutor should be able to monitor student progress 🙂 But this is exactly the kind of information that can help to build relationships and ownership between a learner and content. Obviously this kind of relationship can be even more critical to developing successful learning strategies when using open content without any direct supervision.

Overall I found the day very thought provoking. I was glad to see that a consensus did seem to be emerging that widgets are not some kind of new answer to the age old problem of how we can provide more effective learning opportunities/environments but that they can help with teaching and learning if used appropriately; and if we can all work towards greater interoperability of the basic technology. In this way we can start to provide ways for teachers and learners to access a pic’n’mix of additional, discrete functionality to whichever learning environment they are using be that VLE, webpage, mobile phone etc.

Widget working group start up meeting

The first meeting of our new widgets working group took place last Wednesday at CETIS HQ , University of Bolton. The group has been formed as a response to the widget session at the CETIS conference and part of the day was spent trying to define the groups aims and objectives.

Some broad overarching objectives were identified after the conference, primarily to investigate an infrastructure whereby a collaborative widget server is available to anyone in the H/FE sector – including developing widget server plug-ins for all major web platforms and making widgets easily discoverable and embeddable. Followed by determining models of widget use in teaching practice and support development, sharing, embedding of teaching and learning specific widgets. Whilst these are pretty much in-tact some short terms aims were decided upon, and these were prioritized by the participants.

In terms of infrastructure, two main approaches to widget development and deployment were discussed. These came from the TenCompetence team at University of Bolton and the CARET team, University of Cambridge. The Bolton team have developed their own widget server (called wookie) and have been developing a number of collaborative, shared state widgets (such as chat, voting etc) using (and extending) the W3C widget specification. These can be deployed in a number of systems including Moodle and elgg (see previous blog and David’s post re creating widgets for wookie).

The team from Cambridge on the other hand have taken the shindig/opensocial/ approach and are starting to embed “gadgets” into their new (under development) SAKAI environment. So a short term goal is going to to try and get an instance of wookie server running along side shindig and vice versa and report back on any issues. Over lunch a couple of “lite” interoperability tests showed that widgets built by each team could run in both systems. It was also agreed that some kind of overview briefing paper on widgets (whys, what’s where etc) would be a valuable community output.

Once these initial tasks have been completed we can then look at some of the other issues that arose during the date such as a common widget engine interface, deployment and security of widgets, as well as accessibility, how to write widgets and of course actually using widgets in teaching and learning.

The next meeting is provisionally scheduled for 23rd March in Edinburgh (to tie in with the JISC conference the next day). More information will be available soon and if you are interested please let us know and come along to the meeting.

Widgets, web 2.0 and learning design @ CETIS08

As I’ve already publically declared my love for widgets, I was delighted to co-chair the “planning and designing learning in a world of widgets and web 2.0” with Wilbert Kraan at this years CETIS conference.

Wilbert started the session with an overview of his experiences of building an assessment widget using google docs and sprout builder and integrating it into moodle. The gory details are available from his blog.

One of the many appealing traits of widgets is that they are relatively simple to create and integrate into websites (particularly using wysiswyg builders such as sproutbuilder). However integrating widgets into closed systems such as VLEs can be problematic and generally would require a level of admin rights which most teachers (and learners) are unlikely to have.

This is one area that the TenCompetence project team at the University of Bolton has been investigating. Scott Wilson demoed the Wookie widget server they have built. By adding the wookie plug-in to moodle it is possible to seemlesly integrate widgets to the learning environment. Scott explained how they have been using the W3C widget guidelines to build widgets and also to convert widgets/gadgets from the Apple apps collection to put into wookie and therefore into moodle. This potentially gives teachers (and students) a whole range of additional tools/activities in addition to the standard features that most VLEs come with. A USP of the Bolton project is that they are the first to build collaborative widgets (chat, forums etc) using the W3C guidelines.

Scott made a very salient point when he reminded us that VLEs are very good at allowing us to group people. Most web2.0 tools don’t provide the same ability to group/distinguish between groups of people – there are few distinctions between “friends” if you like. So by creating and/or converting widgets and integrating them with VLEs we can potentially extend functionality at relatively low cost (no system upgrade needed – just add the plug-in) whilst retaining the key elements that VLEs are actually good at e.g. grouping, tracking etc. This is where the learning design element comes into play.

Dai Griffiths, University of Bolton, outlined how the IMS LD specification can provide a way to orchestrate widgets with a set of rules and roles. He showed examples of UoLs (units of learning) which use the chat widget in the wookie server. This is one of the most exciting developments for IMS LD (imho) as it is starting to show how the authoring process can become much easier for the average teacher. You just choose what widget you want to use, decided which students can use it, and when, add other content – widgets -whatever and you have your runnable learning design. I know it’s not quite that simple, but I do think this goes along way to address some of the “lack of easy to use tools” barriers that the learning design community have been facing.

There was a lot of discussion about students creating widgets. Tony Toole told us of one of his projects where they are trying to get students to build widgets. At the moment they don’t seem that keen as they (probably quite rightly) see it as secondary to their “real” learning tasks. However they do like and make use of more informational type widgets which utilise RSS feeds to push out content such as reading lists, course announcements etc. Ross McLarnon from Youthwire showed us the desktop widget application they have just started to roll out to over 130 Universities. Interestingly the university news widget is the most popular so far.

As ever it is practically impossible to summarize the discussions from the session, but some other key issues that arose were authorization. There was a discussion around the emerging oAuth specification which is based on a ticketing system so users don’t have to give username/password details to 3rd parties. It was also agreed that some kind of national educational widget server based on the wookie system would be useful. As well as storing relevant, safe, widgets it could have information/examples of widgets being used in practice and relevant system plug-ins etc.

As a follow on from this session we would like form a working group to explore all these issues in more detail and provide feedback to JISC for potential funding opportunities. If you would like to be involved please let me know either by leaving a comment or by sending me an email (

More details of the session can be found on the wiki.

I love sprouts!

And not just the green ones 🙂 David Sherlock in our Bolton office put me onto Sprout Builder, a very simple widget builder. I have had a play with some other so called simple widget building tools which lost my interest in about 5 minutes or when I realised that they didn’t work with macs, but I have to say this one has really got me hooked.

In about half an hour I had build a widget which displays the outputs for the JISC Design for Learning programme (just taking a feed from the programme delicious site), published it onto the Design for Learning wiki and in my netvibes page. I’ve now just created a widget for my last SIG meeting with audio/video files embedded and a location map which I put into facebook and the CETIS wiki.

Now I’m not claiming that these examples are anything unique, or particularly well designed. However, what I really like about this particular tool is the simplicity of it and the way it integrates services that I use such as rss feeds, maps, polldaddy polls, video, audio etc. Publishing is really straightforward with links to all the main sites such as facebook, beebo, netvibes, pagesflakes, igoogle, blogger . . . the list goes on. You can also make changes on the fly and when you republish it automatically updates all copies.

Tools like this really do put publishing (across multiple platforms/sites) and remixing content into the hands of us non-developers. There are many possibilities for education too, from simple things like creating a widget of a reading list/resources from a delicious feed to a simple countdown for assignments. (OK, that might be a bit scary, but heck a ticking clock works for most of us!). Simple tools like this combined with the widgets that the TenCompetence project are building (and showed at a recent meeting) are really starting to push the boundaries, and show the potential of how we can mix and match content and services to help enhance the teaching and learning experience.