Heart | Soul | Machine

Beyond the LMS


As the Australian delegate, on behalf of the whole continent I say Thanks for Having Me.

This is the context of my institution.

These are our campus locations.

For size and comparison – here’s an overlay of the US.

And I live and work from Wagga Wagga, one of our main campuses.

What I wanted to do today was talk about the Learning Management System, or more importantly what might come after it. But before we get to that I wanted to start with a look at the current state of play.

Over the last decade the LMS has become synonymous with online learning. The LMS has become the default. To many it defines what online leaning looks like, what’s possible and what it’s limitations are.

An element of control is baked into a centralised system. It’s one of its distinct features, but it has some significant side effects. One is an embedded power dynamic that prioritises institutional needs over students, which often reinforces didactic teaching methods where teaching is delivered from a central point.

It’s for that very reason that a key trait of the LMS is a lack of user Autonomy and Agency. Teachers and students lack any real ability to self-govern or to act on their own.

From a central space all can be observed, monitored and tracked. This surveillance is often marketed as “analytics”, and while it may indeed be able to offer some meaningful data it does so at the expense of dialogue and perhaps more importantly permission.

I’d suggest we have reached peak LMS. It has achieved saturation in the market so that is little growth left. For institutions we are now all beholden to innovation being provided to us by vendors and unable to offer substantially different products or services. When everything begins to look and feel the same the return for having an LMS will begin to diminish. Instead of being of value it starts to be a hinderance. If our goal is to create a distinctive curricula and learning experience, then the LMS simply can’t provide that.

There’s a lot of discussion around the concept of the Next Generation Learning Environment. It’s being touted as the solution to the current woes around the LMS.

I am however quite skeptical. You see it’s the same centralised model, with the same inherent problems, the same structures, the same limitations. Sure it’ll be better, faster, stronger – but it won’t change anything.

It won’t create change where it’s needed. It won’t change the visions of what Online Learning looks like.

I’d suggest that the LMS, the default, acts as a container for our thinking. Just like a vessel does with liquid – it shapes the contents. It shapes the performance and what we can imagine is possible. The container provides hard edges, limitations which define how and what we think and do.

So that’s the LMS, but what about our current practices in online learning?

One of the underlying problem is that online learning hasn’t yet adapted to the medium – the web. We’ve simply sought to recreate the physical classroom in the online space. This is what we do with any new medium – radio replicated the theatre, television replicated radio, the early web replicated print. What has happened in the online learning space is a continuation of the trednd – initially it tends to copy the old one, but over time it develops its own distinct form and function.

But online learning has yet to go through that phase. There are examples on the fringes – Downes, Siemens and Cormier’s work on connectivist and rhizomatic learning for example. But for the most part online learning is still an attempt to replicated the Physical Classroom in the medium of the web.

The fact that the biggest LMS company is called “Blackboard” is not a coincidence. Current online learning is an attempt to simply replicate classroom practice. It doesn’t embrace the web. It doesn’t seek to utilise the medium, instead it walls it all off and out.

The LMS also enforces arbitrary Time Boxes which have a profound impact on learning. Access tends to be limited to a session – 6 months – and then students are locked out. Despite the fact that they paid for the learning that occurred there – students no longer have access to the discussions, wikis or content that lives in the LMS. This Time Boxing effectively forces students to start from scratch every session – their profiles, their identity, their network gone. And there’s no way to come back. They can’t return after the fact – to revise, reread, rediscover – it reenforces this concept of learning as a linear processes, all done in step, together and at the same time.

This creates what I’ve called Learning on Rails. Similar to the style of video games where you are immersed in a realistic environment, but have no free movement to explore, simply to complete each task, one after the other. Online Learning tends to consist of linear tasks. Navigating information and working through content is done not as an exploration to hypertext document – but as a series of Next buttons.

So what could the future look like?

If we were looking at the medium itself, what is unique about it? How could we model pedagogies that utilise those traits and features?

Over the past two years working on developing a vision for what online learning could and should be. Based on the large body of educational research that exists, the aim has been to pull together a cohesive model that establishes clear elements to aide the design, development and delivery of online courses.

This is the online Learning model that we developed for CSU. We identified these key elements as part of a curriculum that encourages and enables engagement to occur.

We have also developed the Online Learning Exchange. There you will find a more extensive description of the model. We have also developed up a range of strategies that can be used to help implement the model into courses and used in the design, development and delivery process. The Mixer is a tool to map out practices and how intense their adoption is, the aim to understand that there could and should be variations in the levels depending on the subject and discipline area. It also may help diagnose issues and help direct the design process. The applications area provides case studies and how the model might be adapted over a course, tying in delivery methods and techniques as well as tying in assessments.

This body of work is aimed at moving the university – the largest provider of distance education – into a new way of doing things. It’s a realisation that the while the correspondence model of education works, it has significant inherent problems too. Going online had provided us and many others with a cheaper and easier way, but it also provides an opportunity to rethink what and how we do things.

What our work on the online learning model has uncovered are large gaps between what we want to be able to achieve pedagogically and what the technology, primarily the LMS, is actually capable of.

… but ..

At the moment the LMS is necessary in many cases. It provides a backbone and integrations with administrative functions.

There are issues to do with scale and operations that a Domains program isn’t ready to handle or set up to do. The practical perspective is that if we want to do away with the LMS – then we have to develop a viable alternative.

So let’s look at one way of getting there.

The alternative to the centralised systems, and what I think is the key to changing and transforming what online learning looks like, is moving to a distributed system. The Internet is a Distributed system. It’s success comes from that underlying infrastructure – one that is shared and open. Designed to be more resilient to breakdowns and less vulnerable to attack it also acts to distribute power so it is less abused and better reflects needs of all stakeholders, especially the small and weak.

It’s for this reason that the internet has become a place where emergence happens, fostering more innovation and discovery because people are empowered to do so. They have autonomy and agency within this structure and the ability to carve out and create their own personal and virtual spaces to share.

Domains themselves are a distributed system. They provide each individual a space for autonomous creativity and expression.

They embody these key elements that we want online learning to look like.

But Domains are really only the infrastructure. They provide a mechanism to get things done, but not the method to do achieve it. We have to build – something on top of that solid infrastructure.

My personal journey has been very much focussed on how can we replace the LMS. A couple of years ago a started thinking about how could we do this. What would it look like. And I started with – MYOS. What if we created something like an operating system where we can run our all of our own apps.

And while I think its a sound idea – the reality is that this is too big a task. Recreating applications is a hell of a lot of work, but increasingly we don’t need to do anymore. The main reason is the rise of the API.

Application Programming Interfaces provide a way for different systems and applications to interact with each other. They can share data, send messages to each other and trigger routines to provide different inputs and outputs. APIs are how apps talk to each other. They’ve been part of large applications for a long time but were locked away in the source code. Today though more and more applications are running on the web – and have opened up their APIs. This means developers can link up services – you can cross post social

Last year at the Indie EdTech meetup we started discussing this idea of the Personal API. And having got to spent some time with and listening to Kin Lane I started to realise that you don’t need to run all your own apps. You just need a way to be in control and to coordinate different service and get them to talk to each other. This way if you want to use Dropbox for storage or Amazon S3 – that’s fine. If you want to WordPress or Known – that’s cool too. Utilising a variety of existing web service is actually preferable to building your own, because then it is personal – it’s up to you as an individual to make choices. In this world Agency and Autonomy are baked in.

Utilising APIs at an individual level takes this concept of the web – small pieces loosely joined – and makes it a much more serious proposition. As an individual all these “life bits” are connected – to me. And if I can programmatically control those bit – how they connect and interact, then all of a sudden we have a very new and very powerful tool. At the same time if we utilise existing systems/application then we also have something thats very light weight in terms of development.

So my idea is to develop up a Distributed Learning System. A structured way of utilising a range of technologies to configure a viable alternative to the LMS.

Combining these three key elements we can create a truly Distributed Learning System.

Utilising Domains as the Infrastructre, Applications for functionality and APIs to connect everything together and allow data to pushed and pulled throughout the system.

In this system each student and staff member would have their own System running on their Domain – A Node. Nodes would act as their own entity. They would act as federated points – able to act autonomously but designed so they allow for connections to be made.

And Nodes would connect to Hubs. These could represent subjects or courses – but are much more flexible than that. They could be set up for projects, research, committees and collaboration. Hubs define relationships between Nodes – they allow an agreed set of rules to be developed and define the nature of the relationship – What data will be shared, who with and for how long.

Hubs establish relationships and conventions that allow aggregation and sharing to occur between nodes.

They allow content to be moved around, communication to occur, assessments to be submitted and feedback to be shared.

This map provides a way of understanding how the different components of the system all fit together. The Node acts to manage the Personal side of the equation – connecting apps together and features that are part of the Domains infrastructure.

At the same time the Hubs link together institutional systems and data with those of the nodes. They establish a handshake agreement between parties to ensure that the relationship is negoatiated, data isn’t simply made available and vacuumed up by the institution.

But this isn’t a one-to-one relationship – this is multiple nodes connected. Moving data between students and teachers

Students and teachers would interact with their own node – not the LMS or another site. Their Node. Messages from the hub could be displayed, communications sent back and forth, content would be federated so that students could read, annotate and interact with their materials that they would main them forever. All the tools we currently use for learning would all still work – but in a way that is shaped by the student, and with them having access and choice.

The radical of this that students would be able to take their learning with them wherever they please – and for as long as they please. They wouldn’t be chained or confined by the institutions any more. We could start to see this concept of lifelong learning actually be supported by the technology. Students would retain copies of their learning and they would be able to use it how they wish. Creating eportfolios, showreels, blog posts – reusing and repurposing their learning.

At the moment Domain of Ones Own relies on co-opting open source applications to achieve certain needs. Blogging tools like WordPress have become powerful tools in the hands of skilled artisans, but outside of those few individuals have we gone much further than simply blogging? But what if we used those applications the infrastructure of Domain of Ones Own to develop tools specifically for learning?

I still refer to this post from Andrew Rikard when thinking about domains and students. One way of the thinking of the Distributed Learning System is to provide a way to unbudle the students learning from their domain and digital identity. By using APIs we can be more focussed on who and how we share content. And in doing to it not only provides a safe space for students to work in – and still engage with the web, but to do it on their terms, while maintaining their authority and avoiding turning domains into the next checkbox assessment.

The real potential of the DLS is the development of new applications and tools. To utilise the latest technologies and the open nature of the web and software to create new applications that focus on learning, on a pedagogy of the web. We develop methods of learning that are of the web, and are based on discovery, exploration, creativity and reflection.

At the moment there are some fantastic innovations happening out there – over this conference we’ve been introduced to just some of them. But we’re dealing with blips at the moment. To improve and make a difference to online learning and perhaps education as a whole, we have to work to share our knowledge and experiences. We also need to make it easy to adopt them. To learn from one another and our experiences. By creating a system, and while that word may scare some people, it’s really just a way of working, we could share more easily. APIs provide a way that we could share, adopt and adapt new developments more easily – between users and institutions. Having some commonality would provide some cohesion

Moving into the future – if we want to develop and deliver a truly distinctive curricula and learning experience. One that I think would produce the kind of graduates that would thrive in the future. A distributed learning system radically changes the possibilities and provides a way to really develop self directed learner. Providing students with a level of autonomy and agency that is simply not possible within in the LMS and centralised systems, they will develop the skills to manage and define their own learning in a life long way.

If you have comments, questions or ideas – let me know. I’d love to hear them and bounce ideas around.