I am investigating the connections between Informal Learning, Communities of Practice, Network of Practice and Connectivism. Found an interesting conversation between Jay Cross, George Siemens, Dave Cormier and (on chat) Stephen Downes on EdTechTalk#23, Nov 3, 2005.

Found also Wenger’s interview and lecture at a KnowledgeLabs e-portfolio Konference where he talks about learning as meaning/sense making.


What inherently constitutes a connectivist learning ecology? What specifically differentiates it from a collaborative, Web 2.0 or informal learning enabled learning environment? Was the CCK08 course representative of the Connectivist learning ecology?

Lisa Lane wrote a list of recommendations on the CCK08 experience. Bradley Shoebottom has devised his own structure. I proposed the concept of Network Based Training. There are many others.

George Siemens writes:

I like the idea of thinning our classroom walls and allowing connections to be formed between concepts from other subject areas. But that responsibility shouldn’t rest on the educator. “Getting on the same page” (author’s words) seems a bit at odds with opening up class rooms. We need to all get on our own page, form our own connections, our own understanding of different fields. It seems that the desire still runs high for educators to apply increased organization when problems become intractable. What is really needed is a complete letting go of our organization schemes and open concepts up to the self/participatory/chaotic sensemaking processes that flourish in online environments.

Monty Paul ties in connectivism into social constructivism.

The idea of connectivism (Drexler, 2008 ) ties in well with social constructivism, demonstrating how new generation learners use the power of our networked world to tap into remote sources of knowledge, including experts in various fields. These learners work in a world without boundaries from a technological point of view. They are adept at finding, storing, managing and sharing information using new web-based applications. More importantly, they are involved in knowledge creation, using blogs, wikis and other on-line applications to mash and developing new ways of looking at and using information. These students bring fresh challenges for learning institutions across the educational spectrum, given their need for a fast moving, game oriented learning (Pensky, 2001) which traditional learning environments are hard pressed to provide.

But I want to discuss the difference between the creation of an ecological blueprint (if there could be one) that “allows” connective learning and what would constitute an ecological blueprint that “is” inherently a connective learning design/blueprint. For example, the difference between saying “the hotel lounge is Wifi-enabled” is different than saying that “I can check my email in the hotel lounge”. After all, it’s the conversation rather than the blogging tool that’s more important, right?

George Siemens contrasts behaviourism, cognitivism, constructivism and connectivism in the light of Ertmer’s and Newby’s five definitive questions to distinguish different learning theories. For connectivism, he states:

  1. How does learning occur? – Distributed within a network, social, technologically enhanced, recognizing and interpreting patterns
  2. What factors influence learning? – Diversity of network, strength of ties
  3. What is the role of memory? – Adaptive patterns, representative of current state, existing in networks
  4. How does transfer occur? – Connecting to (adding) nodes
  5. What types of learning are best explained by this theory? – Complex learning, rapid changing core, diverse knowledge sources.

(George’s responses in italics)

George and Stephen also talk of the impact of chaos theory, self organization and complexity on the learning process. They also refer to the impact of this way of learning on traditional notions of power, control, validity and authority (among others). So what would constitute the learning ecology that is connective? It should be one that inherently:

  1. Enables us to recognize and interpret patterns that exist (way finding, sense making) ; indeed, generate our own new patterns
  2. Helps us build adaptively on and capture existing patterns given a rapid changing core and diverse knowledge sources
  3. Provides a distributed environment (both for knowledge and people)
  4. Provides avenues for social collaboration
  5. Is technologically enhanced to deal with diverse processes/circumstances such as negotiating information overload, self organization, determining order within chaos etc.
  6. Enables us to leverage and expand on a network that is diverse
  7. Helps us build ties at varying strengths that in turn may determine the efficacy/effectiveness of our learning
  8. Enables us to negotiate complex learning needs

Replace “what would constitute a connective learning ecology?” with “what kind of educator would suit or engender a connective learning ecology?” and it becomes easier to think about the problem instantly.

That is, the answer that the educator should “model” and “demonstrate” his connective learning process/ability/efficiency while the learner should “practice” and “reflect” (I think “observation” and “experimentation” are equally critical skills), makes sense because the ability to do all of the above needs to be learnt by the learner. The objective is perhaps that the learner be empowered with the learning skills and ecologies of the educator (as George Siemens says “…A curator is an expert learner”).

What if there was no educator or formal role for one? What happens in that truly open, autonomous, distributed, uncontrolled network? Is there an ecology for the solitary learner; for the ones that are faced with unequal access; those who have technological/social barriers or limitations?

In a sense then, perhaps we should look at the design and metrics of a connectivist ecology from a different lens altogether – where the ecology contains components that inherently propel the learner to become a curator.

Instead of providing a chat tool and a structured interaction and participation schedule, it should provide (for lack of a better technologically unchallenged term) a “default” mechanism for learning that propels the learner to make connections, practice and reflect, observe & build & recognize distributed knowledge patterns.

It is here that the discussion around types of networks becomes really important. At the neural level, it is really immersion into the environment (“increased awareness?”); at the conceptual level it is the ability for the ecology to provide some ways of exploring and building concept patterns and at the social level, the learning network of people (and devices) itself in a given context. It is here also that we should perhaps attack the concerns around motivation and participation.

Perhaps when the three (and there may be more) types of networks come together in some way, they become really powerful for learning. For example, experiencing rain-drops, recognizing the dark cloud visual and listening to the thunderclap, associating it with concepts of cloud formation and effects of rainfall, and, warning your friend not to venture out, may be an example of learning could manifest itself given this three way association (there could be self spiraling associations within a network type itself).

Where would the metrics then come in and how would they be designed? In another corporate context, I once read a powerful article by John R. Hauser and Gerald M. Katz titled “Metrics: You are what you measure!”. In my mind and as they state, successful metrics are good if the actions and decisions which improve the metrics also improve the firm’s (read “learner’s”) desired long term outcomes (read “learning ability” or “expertise”). They list seven pitfalls of metric design and how these can completely subvert the metric design exercise. They also list an equal number of steps to design good metrics such as “Listen to the customer” and “Understand the inter-relationships” all of which I think are useful ways to think about what to avoid and what to follow.

The main point is that we need to understand if score, time elapsed, distance between two nodes (a.k.a. social network analysis), e-portfolio submission & ratings et al are good metrics in this connectivist ecology. Instead, wouldn’t we ask questions relating to or perform investigation into how well the learner is able to learn using the “default” mechanism I referred to earlier? For example, speed of learning could be perhaps (or maybe I am being too simplistic) the rate of change of new patterns, network connections, conversations; or the measure of expertise would be the number, qualitative rating, network perception or rate of interaction between you and the resources in your network?

As always, would love to be corrected and to know your thoughts!

I have an occasion to do a little research on the Montessori method. Named after Dr. Maria Montessori, who, in 1896, was the first woman in Italy to graduate out of medical school, the Montessori method seems to have rich similarities with Connectivism.

The basic Montessori concepts are pretty well known by now (Montessori in Perspective, 1966). 1 – The teacher must pay attention to the child, rather than the child paying attention to the teacher. 2 – The child proceeds at his own pace in an environment controlled to provide means of learning. 3 – Imaginative teaching materials are the heart of the process. 4 – Each of them is self-correcting, thus enabling the child to proceed at his own pace and see his own mistakes. If you were to look inside a Montessori classroom, you would get the impression of “controlled chaos” because each child would be quietly working at his private encounter with whatever learning task he or she chose (Montessori in Perspective, 1966).

(Quoted from http://www.webster.edu/~woolflm/montessori2.html, retrieved January 10, 2009)

Dr. Montessori said “I studied my children, and they taught me how to teach them.” And also her general principle – “first education of the senses, then education of the intellect”.

‘The essential thing is for the task to arouse such an interest that it engages the child’s whole personality’ (Maria Montessori – The Absorbent Mind: 206).

This connected with a further element in the Montessori programme – decentring the teacher. The teacher was the ‘keeper’ of the environment. While children got on with their activities the task was to observe and to intervene from the periphery. (Here there are a number of parallels with Dewey).

(Quoted from http://www.infed.org/thinkers/et-mont.htm, retrieved January 10, 2009)

Cammy Bean, was at the same juncture as I am today. Sending my son to formal school. But my son also has had an year of so already under his belt at a Montessori pre-school. She asks:

So, the question is, can we make the Montessori Method a part of the coporate learning environment? Is that what all this informal/DIY/learning 2.0 stuff is all about?

Constance Steinkuehler in her post Digital Montessori for Big Kids, likens virtual worlds to digital environments natively conducive to the Montessori method.

The Microlearning conference 2005 proceedings“Learning and Working in New Media Environments” (June 23-24, 2005, University of Innsbruck, Austria) has an interesting paper by Gernot Tscherteu titled “The Blogosphere Map – Visualizing Microcontent Dissemination” (looks very similar to what Valdis Krebs does :)). The principles of the Montessori method are referenced in the context of microlearning:

  • the Montessori pedagogue is acting in the background, and,
  • Learning by playing in mixed groups
  • Free schedules, no collective teaching
  • Prepared environment
  • Learning materials are kind of small interactive games for real life learning experiences
  • Sensomotoric

Very similar to what I have learnt about connectivism e.g. the second is a characteristic of networks if you think about it.

Microlearning itself seems to be an interesting concept. Theo Hug summaries different dimensions together in his paper “Microlearning: A New Pedagogical Challenge” in the same Austrian conference.

There is not one precise definition which covers all the different concepts. In my view there are versions which are brought forth by different interpretations of particular dimensions such as:

  • Time: relatively short effort, operating expense, degree of time consumption, measurable time, subjective time, etc.
  • Content: small or very small units, narrow topics, rather simplex issues, etc.
  • Curriculum: part of curricular setting, parts of modules, elements of informal learning, etc.
  • Form: fragments, facets, episodes, „knowledge nuggets“, skill elements, etc.
  • Process: separate, concomitant or actual, situated or integrated activities, iterative method, attention management, awareness (getting into or being in a process), etc.
  • Mediality: face-to-face, mono-media vs. multi-media, (inter-)mediated, information objects or learning objects, symbolic value, cultural capital, etc.
  • Learning type: repetitive, activist, reflective, pragmatist, conceptionalist, constructivist, connectivist, behaviourist, learning by example, task or exercise, goal- or problem-oriented, „along the way“, action learning, classroom learning, corporate learning, conscious vs. unconscious, etc.

Interesting meanderings so far!

Is there something like that at all? In a discussion yesterday, an important point was made by a participant – we don’t want perfect environments to be created for our learners, even if we could create them.

Why is this important to discuss? Everywhere around us there are “frictional” forces that impede or obscure – could be authority, access, lack of infrastructure or others – the learning process.

The ability to learn to cope with these forces becomes equally critical as the process of sense-making or wayfinding in a connectivist paradigm.

What is this ability?  The best way to place this ability in stark contrast is to assume a limiting factor. Let us say the individual has no access to (say) Web 2.0 technology. Specifically, the ability to form online networks / inter-personal relationships and instant online collaboration does not exist for this individual.

For her, sense-making would be based in a world of books & letters, local resources, chance encounters and possibly luck in tems of finding the right connections for her purpose. She would then possibly compensate for this frictional force in many other ways and an important factor here would be individual agency, apart from environmental facilitation and personal skill. She would actively seek and pursue opportunities that allow her to overcome this frictional force in a unique manner.

This ability to innovate & learn within physical world and personal constraints is equally important as the process of learning itself – maybe an inseparable aspect.

Connectivism makes the negotiation of information an important aspect of the learning ability, maybe it should include negotiation of real world constraints as well.

In a previous post, I tried to identify some of the impacts of connectivist practice on visual design. Primarily, these are:

  • Usability
  • Visual “languages”
  • HCI Design
  • Programmable patterns depicted visually
  • Shared visual patterns
  • Ease of authoring new media and media mashups

I was referred to ManyEyes through a blog post (I think it was George’s post) and found it extremely interesting, not because of what it does (because that has been experimented earlier), but because of the way they have put it together – large number of visualization types and ease of authoring.

What is especially interesting is what they call topic hubs. What this means is that anyone can go in, start a discussion topic and add visualizations and data to it.

So these are really combinations of two different ideas – mashup between data and presentation style, and, collaboration around a shared object(s).

I think these are powerful ways of visual collaboration. Within connectivism, they offer an important way of making connections thereby impacting learning. 

What would be even more interesting is if someone took two or more different media mashups and started mashing them together.

For example, a world map showing pollution levels across major cities could be drilled down (or linked to other related visualizations) into a bar chart which could then be tagged to a Technorati tag list in addition to a Twitter conversation in addition to….

For the visual designer (as well as the educator), these represent important starting points to think about the multi-dimensionality provided by connectivism.

Outliers, Gladwell’s 2008 book (and I have not read his earlier work yet), is something that I started on yesterday. It has caught my attention from page one.  Gladwell wants us to

…appreciate the idea that values of the world we inhabit and the people we surround ourselves with have a profound effect on who we are.

Gladwell explores success (“Why do some people achieve so much more than others? Can they lie so far outside the ordinary? What is the secret of their success?”). These questions directly attack our notions of talent and expertise. And in many ways bring out the effect systemic decisions may have on development of this talent and expertise – somewhat chaotic, sensitive to initial conditions.

In his chapter on the “Matthew Effect” (Matthew 25:29; term coined by sociologist Robert Merton), Gladwell does an interesting analysis. He took the birth months of junior hockey league players in Canada and found that most of them were born between the first three months of the year. Having found this, he went on to study other such teams, the US non-school Baseball league, European soccer, the Czech National Junior soccer team and the studies by economists Kelly Bedard and Elizabeth Dhuey on the relationships between TIMSS (Trends in International Mathematics and Science Studies) scores and birth month. He found similar patterns in the data!

Why? Apparently because, each of these had a system defined cut-off date for eligibility. In the Canadian Junior Hockey leagues, the cut-off date is January 1. So a player reaching the age of 10 on Jan 2, could be playing alongside someone who would reach that age perhaps in December that year – a huge difference at the age of 10, would you say?

Similarly, in TIMSS, among the fourth graders, the children who were the oldest scored at least 4-12 percentile points better than the younger children!

So practically speaking, if you group by “ability” and put older (“more mature”) students in advanced streams where they are given better opportunities to learn, they then have an advantage that is iteratively increased as they move from grade to grade.

Huh! Dubner and Levitt (Freakonomics) would be pretty kicked to see this research 🙂 .

The other aspect is that the kids who were born in the later months got lesser and lesser attention or opportunity (very few Czech soccer players born between July-December, for example), which meant that a large population of students did not make it because of when  they were born?

So what does it mean to have innate talent or intelligence then? What it does it mean to be an expert? Gladwell quotes neurologist Daniel Levitin who states:

The emerging picture from these studies is that ten thousand hours of practice is required to achieve the level of mastery associated with being a world-class expert – in anything. In study after study, of composers, basketball players, fiction writers, Ice skaters, concert pianists, chess players, master criminals, and what have you, this number comes up again and again. Ten thousand hours is equivalent to roughly three hours a day, or twenty hours a week, of practice over ten years. Of course, this doesn’t address why some people don’t seem to get anywhere when they practice, and why some people get more out of their practice sessions than others. but no one has yet found a case in which true world-class expertise was accomplished in less time. It seems that it takes the brain this long to assimilate all that it needs to know and achieve true mastery.

(also referred at PBDB).

That’s interesting because, where do you get the opportunity (and where is it systematically denied?) to practice twenty hours a week over 10 years if the analysis is true at all.

Gladwell goes on to look at Bill Joy, the Beatles, Bill Gates and many more. The refrain is that opportunity to succeed is too important a factor to consider in your success. And that, intuitively, strikes more than a single chord in me personally and because of what I have experienced in CCK08.

Open, accessible networks may provide our children what they need to be successful. Their ability to form connections and uncover opportunities for individual learning and growth may result in covering some of the great disparities caused by existing world structures, be they economic, educational, legal or others.


Starting with mind maps as a central way of modeling thoughts, the tool is quite like CMap. Allows tagged links and resources from delicious and other tools to be used and attached to each concept or thought node. It also allows “souping” knowledge using a centralized repository on the web that you can host with permission. However, this tool goes beyond in terms of usability and features. First, it is Web 2.0 based. Second it allows fragments or entire concept maps to be “imported” from one user to another. They call these concept maps journeys that can be shared and embedded.

Interesting concept because our project group in CCK08 was trying to get our maps together but could not do so because of disconnected terminology – no way to map nodes and compare because of language or representation (two different terms meaning the same, even typos, inconsistent use of mapping rules etc) problems. IMINDI does not address those problems.

They also have something called a Mindex or mind index which is really putting together the different journeys starting from a single keyword. So they take a keyword/concept term and check their database for all first level associations and so on. The results are pretty powerful at first glance. When quizzed on the business model, the CEO of IMINDI remarked that unlike Google that has to write a search engine to collect thoughts (read “nodes”), they are getting their users to pubish and voluntarily make the connections with what they want to know, some kind of reverse search (if there is such a thing).

The reviewers thought this was, well, a little far fetched and could not relate it to how people would even want to invest time upfront to create a mental map in the first place. They  thought the company would do better to focus on the enterprise as a testing ground.

This is an example of a good thought, but no clear articulation of value for the effort.

I read George’s post and reference to CISCO’s beliefs with great interest. There are systemic barriers to change almost anywhere change occurs. But John Chambers suggests that changing the structure of the organization is key to making innovation work somewhat like C K Prahalad and M S Krishnan outline in their book, The New Age of Innovation. They talk about co-created experiences and access (rather than ownership) of global resources as two fundamental pillars of a organizational structure. 

The locus of value is seen to be shifting from products and services to experiences that are personalizable at a large scale, yet being affordable and high quality. N=1 involves a new approach to access and use of resources. The authors term this R=G. The necessity is that no one firm can even attempt to own/access all the different resources that it would need for creating new experiences for the customer in an N=1 world.

To this end, they build a framework for innovation that organizations can leverage. This seems to be a useful approach for looking at how educational systems can be transformed by structural change in the pursuit of innovation and excellence. It also seems consonant with Connectivism as the inspiration and rational basis for the change to occur.

As I have started feeling, maybe the change should percolate from top down rather than from K-12 to adult education. But a thorough analysis of and architecture for the structural changes that may be required seems to be crucial at this moment.

I read a series of contributions by Stephen, George, Pontydysgu, Attwell, and reviewed PLE diagrams and Wiki entries. George makes the point that PLEs are antithetical to existing educational systems, which are really  structures of power, accountability and control based in a sociological context, not focussed on learner needs and goals.

For this reason, PLEs, which are based on learners needs/goals and concerned with individual and personal autonomy and learning, cannot move into the center of the learning process until the underlying power relationships change.

Attwell suggests that the next steps in research and development of PLEs should include

  • examining in depth how individuals are using computers for learning in different settings (especially non-educational technology) and outside the setting of formal educational programmes;
  • exploring the relationship between informal learning and formal learning in developing competence;
  • examining different forms of competence and how educational technology can support such competences;
  • examining the use of different social software applications for learning;
  • examining in depth the nature and form of computer mediated interactions between learners in different communities;
  • examining the implications of persistence of data for Personal Learning Environments;
  • examining the different ways in which learners might wish to represent learning (both formal and informal);
  • examining what materials are used for informal learning and how they are used;
  • exploring the implications of changing forms and patterns of learning for educational institutions;
  • exploring ways of representing and patterning learning activities interactions;
  • exploring ways of utilising different services – both within and between institutions and with broader communities – to support PLE-type activities;
  • exploring issues in standards and interoperability to facilitate PLE-type development;
  • exploring how PLE-type applications and services can be integrated or work alongside existing educational applications and services

These are important inclusions. For example, Attwell talks about standards, interoperability, nature and forms of computer mediated interaction and data persistence, which are all going to be important for toolmakers and designers of PLEs.

Pontydysgu talks about the inherent contradictions in capitalist societies (and the associated systematized superstructures of education systems) and the power of individual & collective agency (as a transformative agent).

It is also important to reference Lisa Lane’s strategy for change in this context:

The solution is subversive application of connectivist and other useful learning ideas within the current structure, an insurgence for the purpose of fostering emergence.

The discussion begs two questions:

  1. Apart from shifts in underlying power relations, are there any other enabling factors for the adoption of PLEs (as tool/process/concept)?
  2. Is the mode of change that is required “insurgent” or “revolutionary”?

This is the final project submission for CCK08. Thanks to George and Stephen for providing an extremely effective ecology for Connectivism and connective knowledge! Thanks also to all the participants who made this an enjoyable and thought provoking experience. Hoping that we will continue this dialogue onwards into the future.

YouTube video accompanying this project submission.

A special note for the musketeers – Carlos and Maru – thank you for your support!

 The questions that were asked of us and my responses are given below.

1. What is the quality of my learning networks: diversity, depth, how connected am I?

Extremely connected. Although I was unable to contribute as comments as frequently as I would like in the Moodle forum or CCK08 blogs, I think I managed to read/skim through a very large number of the forum and blog posts with the help of the subscriptions and the Daily. Maybe I need to be more involved in ideas that others propose to be able to interest others in mine!

However as far as learning goes, I think the blog posts themselves, rather than the sparse conversations through comments around most of them, served more as thought triggers. The Moodle discussions were real conversations though that provided a far greater diversity and depth. Diversity was immense though I did feel the lack of interaction with profiles similar to my own.

I think there was a lot of depth across all the weeks of discussion that the participants exhibited. This depth is really what made me understand the gaps in my own education. The biggest decisions I had to make was to balance the travails of running a company with the ambitions I had to transcend these gaps.

2. How has this course influenced my view of the process of learning (assuming, of course, that it has)?

This course, for me, actually started a few months before it did physically. I had been following up with George and Stephen’s writings and debates all over the web (started with Learning 2.0) and was very impressed and in-line with what I read and heard.

But for me, the real impact was not learning how connectivism was defined, was distinct and was impacted by a confluence of many theories/concepts, but the basic understanding that knowledge could really be connective and that learning could be the process of making connections.

I struggled with the technology bias because I usually resist that in whatever I do, despite being a technologist, but I think I am at peace now with the way George and Stephen approach it (especially from the last Friday session of the course).

This course has also exposed a number of factors to me that can influence (both engender and impede) connective learning, but has not yet gone more beyond the use of technology per se in the implementational aspect.

3. What types of questions are still outstanding?

The types of questions that are still outstanding are around:

  • Neuroscience – can the new developments really describe intelligence?
  • How would connectivism work in the strict absence of technology or language or even people?
  • What other theories are there that could impact connectivism as strongly as self organization, chaos and complexity?
  • What are *acceptable* practical ways of implementing the theory?
  • What kind of a perspective do toolmakers need to adopt when crafting PLEs or the next generation of tools?
  • What kind of open, autonomous ecologies can replace the structures of traditional classrooms – are these ecologies going to be sustainable and extensible?

Many more questions are more domain specific. I think each node in my concept map has covered an entire very large domain of research and thought that I need to explore in a structured manner.

4. How can you incorporate connectivist principles in your design and delivery of learning?

At this point, I am dealing mostly with corporate training. I can see how these could be applied, but the predominant question (and it comes from within the dominant paradigm) is how progress can be assessed and measured using designs based on connectivist principles.

With the work we (Servitium) are doing in simulations, we have taken a step away from the semantic web / RDBMS representation of knowledge to associative knowledge. We are using loosely connected “fragments” to create “patterns” of knowledge, but still need to research how the latest techniques in neurosciences (and AI) can help make these more explicit and sophisticated.

Another initiative we have taken is to make the design process itself connective through sharing and collaboration using blogs of training design in a way that is accessible by a variety of stakeholders. The results of that one are still to be obtained.

RSS CCK08 Yahoo Pipe

  • An error has occurred; the feed is probably down. Try again later.
%d bloggers like this: