Page 1
PrePrint
Jon Dron and Terry Anderson (2016) The Future of E-learning. In the SAGE Handbook of E-learning Research (2016) Second Edition. Edited by Caroline Haythornthwaite, Richard Andrews, Jude Fransman and Eric M. Meyers. Sage
The Future of E-learning
INTRODUCTION
This is not the first attempt to predict the future of e-learning and our first confident
prediction is that it will not be the last. Our intent in this chapter is to focus less on the digital
technologies involved and more on broad trends and consequences, especially as they affect
and are affected by the pedagogies and their surrounding educational infrastructures. We do
not wish to predict the future so much as to characterize its general form and examine the
implications for the present and the futures that emerge.
E-LEARNING
The term ‘e-learning’ was independently invented several times in the mid-1990s (Cross,
2008): ‘it was in the air’. Twenty years on, the term is beginning to sound like a quaint
anachronism. Before it dies out altogether, we will provide a definition so that future readers
may better understand the context of which we write: E-learning is a combination of
methods, structures and networked electronic tools orchestrated into systems that bring
about, or are intended to bring about, learning.
Inherent in this definition is that part of the orchestrated assembly must include methods
of teaching or, more concisely, pedagogies. From this perspective, pedagogies are
technologies (Dron, 2009). Like all technologies, no one part exists in isolation. All are
mutually constitutive and dependent on one another – parts of a single orchestration (Arthur,
1
Page 2
2009) – and so all must fit together. The ‘e’ part of the term therefore matters as much as the
‘learning’ part. Neither occupies a privileged place, any more than the engine of a car is any
more or any less important than the skills of its driver. The orchestration matters as much as
does what is orchestrated.
The role of the orchestrator in e-learning is always distributed – programmed machines
and people such as teachers and textbook authors may do some of it, but learners themselves
always do at least as much. So too do the myriad of people that either directly or indirectly
contribute to the orchestration through direct interaction or the design of tools, content and
processes.
E-TEACHING
With the foregoing in mind, we make the distinction between e-learning and e-teaching. The
vast majority of uses of the term ‘e-learning’ in the literature actually describe e-teaching,
where the focus is on teaching interventions rather than directly on the learning that may
result (Vermunt & Verloop, 1999). Human learning, whether construed as a socially
distributed activity or an internal psychological or physiological event, is never electronic.
Moreover, professional teachers and learning designers have only a minor part to play in
most e-learning. In 1999, John Chambers, then head of Cisco Systems, was mocked for
predicting that Internet traffic caused by education over the Internet would make email usage
seem like a rounding error (Neal, 2007). He was, however, substantially correct. Almost
everything shared on the Internet, almost every interaction, is an opportunity for learning,
whether deliberately sought (e.g. Google, Wikipedia, MOOCs, YouTube) or as a side-effect
of interaction (e.g. Facebook, email, Twitter). Whether finding or passing on information,
challenging beliefs or affirming them, inciting critique or inspiring ideas, our reified online
interactions are learning interactions. The Internet has transformed in both scale and scope
2
Page 3
the sharing and the collective construction of knowledge – thus exploding the opportunities
for learning. On the Internet, almost everyone is a teacher and everyone is a learner, whether
intentionally, effectively, accurately, reliably or not. Explicit e-teaching only scratches the
surface of e-learning’s impact. Before we explore the consequences of that, we consider a
little of the history of e-learning futures, to help explain where we are now and to provide
useful boundaries that allow us to shape our enquiry.
THE EDINBURGH SCENARIOS AND WHAT THEY TELL US
In 2004 what was, for a field named barely ten years previously, grandiosely described as the
Largest Scenario Planning Exercise ever undertaken in the e-learning community, the
Edinburgh Scenarios (Bell & Stewart, 2004) resulted in a spectrum of four possible broad
scenarios for the future of e-learning (Figure 26.1):
Virtually vanilla – technology to support e-learning mainly in the hands of
large corporations and institutions
Back to the future – trust lost in e-learning, with a subsequent return to
traditional values and face-to-face teaching
Web of confidence – the Web enables people to learn and work together in new
ways, leading to decentralization and a shift in power away from large
organizations
U choose – a world where people are frustrated by and reject new technologies
but find new ways to gain greater control over their learning and greater
independence from central authorities.
The scenarios were designed to be relatively neutral with regard to any specific electronic
technology. As we write this, more than 10 years on, it is informative to consider how they
have stood the test of time.
3
Page 4
[TS: Insert Figure 26.1 here]
Figure 26.1 The Edinburgh Scenarios (after Bell & Stewart, 2004)
Virtually vanilla
The ‘virtually vanilla’ scenario is in full swing and stronger than ever. There is greater e-
learning centralization than there was ten years ago, whether in the form of instructivist
MOOCs, moves by publishers to take greater control of institutional learning, or the ubiquity
of learning management systems (LMSs) in both higher education and schools. The likes of
Blackboard and EdX are thriving as never before, replicating models of learning and teaching
that stretch back at least as far as medieval times; automating the well-entrenched process
models of most traditional institutions and thereby increasing access and market share of
these large organizations.
Back to the future
Although formal online learning is firmly established, with at least 32% of students in the
USA alone now taking at least one fully online course (Allen & Seaman, 2013, p. 4) and the
vast majority of all courses incorporating some online elements, the majority of students still
prefer face-to-face over online learning (Taylor et al., 2011). If it has achieved nothing else,
the popularity of MOOCs has renewed a focus on the strengths of traditional face-to-face
education (Ritzer, 2013) and traditionalists have found a new assertiveness about the value of
older pedagogies (Brooks, 2012).
Web of confidence
The ‘web of confidence’ is utterly pervasive. Technologies such as Google Search,
Wikipedia, Facebook, blogs, Twitter, Stack Overflow, the Khan Academy and Reddit have
come to dominate informal learning, largely replacing books, newspapers and journals for a
growing segment of the population, as well as making ever larger inroads into the formal
4
Page 5
learning space, whether or not this is condoned by teachers. The first and still popular format
of MOOCs to share the name arose out of Connectivist theory (Siemens, 2005), relying on
decentralized, networked, learner-owned technologies that exemplify the web of confidence.
Technologies used for learning are cheaper and more distributed than ever before and are
increasingly learner-controlled.
U choose
The ‘U choose’ scenario, in which learners engage in physical spaces independently of
traditional institutions, is also thriving, growing in strength and reach. Trends like the Maker
Movement have led to an abundance of small local workshops and local events, while ‘Free
Universities’ that offer courses provided by enthusiasts in local settings, such as
http://freeuniversitybrighton.org, are trending, and the range of topics and contexts around
which people are ‘meeting-up’, from TEDx to book clubs, is expanding everywhere. Most
provide at least a website and other digital technologies to coordinate their activities, but the
actual learning activities themselves are local and face to face. Similarly, the home-schooling
movement has grown considerably, notably in the USA: for example, it experienced a 17%
growth rate in the period from 2007 to 2012 (NCES, 2013) and now accounts for over 1.78
million students. While this growth is only possible because of the widespread availability of
content and interaction online, much of the learning activity is both beyond institutions and
based on the physical co-presence of parents and children.
DIVERSITY AND DIVERGENCE
What the intervening decade between now and the formulation of the Edinburgh Scenarios
teaches us is that there is no single future of e-learning. If it can happen, it probably will. At
least as significantly, this does not normally cause the extinction of what came before. As
Kelly persuasively argues, old technologies seldom if ever die (Kelly, 2010). Moreover,
5
Page 6
existing patterns become a part of the new and ever expanding tapestry. Technologies evolve,
as Arthur (2009) shows, through a process of assembly and recombination. Each innovation
builds on the last and, in turn, opens up new possibilities for further innovation as well as
new constraints and dependencies. The more technologies that are available, the more they
may be assembled into new ones. Kauffman (2000) explains this as an inevitable expansion
of the adjacent possible: every new change makes possible further changes which, in the
absence of catastrophic disaster, cannot but lead to greater diversity and complexity.
We can thus expect all of the Edinburgh Scenarios to persist and spawn new forms that
may not currently exist or, more likely, that already exist in nascent forms for, as Gibson
observed, the future is already here, just unevenly distributed (Rosenberg, 1992). No one
scenario will dominate forever because each will find its niches. It also depends on where one
chooses to draw the system boundaries. For example, while educational institutions have long
been in the thrall of the makers of learning management systems, with a strong centralizing
‘virtually vanilla’ trend, students (and even some teachers) at such institutions often embrace
the ‘web of confidence’, making extensive use of personal learning networks supported by
the likes of Facebook, Twitter and Whatsapp, not to mention learning-support sites like
Wikipedia and StackOverflow, Learni.st and Google Search.
THREE GENERATIONS OF DISTANCE EDUCATION
PEDAGOGIES
Diversity characterizes the future of online pedagogies as much as it does organizational
forms. We have previously described distance learning (and, by extension, e-learning) in
terms of generations of pedagogies (Anderson & Dron, 2011). Up to now we have observed
three generations: the behaviourist/cognitivist, the social constructivist and the connectivist.
In keeping with Kelly’s observations on the immortality of technologies, the emergence of
6
Page 7
each new generation has not replaced the last, but coexists with and extends it. And, in
keeping with Arthur’s perspective on technologies as assemblies, the dominance of each
pedagogical norm is closely interwoven with the technological capabilities that it employs.
Generation 1: cognitivist/behaviourist (instructivist) pedagogies
Before cheap and ubiquitous many-to-many technologies like forums, newsgroups, email and
learning management systems, it was not economically viable to attempt many-to-many e-
learning, so the first generation of e-learning was necessarily concerned with individual-
oriented methods of teaching and training, employing behaviourist or cognitivist theories that
address the efficient transfer of knowledge to individual learners. It is characterized by well-
defined learning outcomes and an objectives-driven approach. The generation began by
assembling the pedagogical methods of books and written correspondence with multimedia
and computational features of computers, first through computer-based training, CD-ROMs
and, later, through web-based training through to modern instructivist MOOCs. The
interactivity of electronic media provides opportunities to simulate, stimulate and offer
adaptivity and feedback, but cognitivist/behaviourist methods are mostly premised on the
notion of knowledge transfer – teachers primarily provide, rather than nurture, knowledge.
Such approaches mainly fit firmly within the ‘virtually vanilla’ quadrant of the Edinburgh
Scenarios.
Generation 2: social constructivist pedagogies
Starting with PLATO in the 1960s through to bulletin boards and MOOs in the 1980s,
pioneers experimented with many-to-many pedagogies, but cost, technical limitations and
complexity limited their impact. As the Web grew in the 1990s, new many-to-many tools
emerged, at very low cost and with relatively high ease of use. Social interaction at last
became straightforward and reliable to achieve and social constructivist approaches came to
7
Page 8
dominate. Following already popular precepts from earlier thinkers like Vygotsky (1978) and
Dewey (1916), these enabled more open, unpredictable, learner-focused approaches to the
social construction of richer knowledge through engagement, interaction and collaboration.
Through the 1990s and 2000s, this was assisted by the growth of tools such as learning
management systems that provided simple and usable forums and collaborative workspaces.
While far more open, flexible and accommodating of different learner needs, these social
constructivist approaches still assumed a set of group goals, set by teachers: they were about
intentional, guided learning. Thus, they also mainly sit within the ‘virtually vanilla’ quadrant
of the Edinburgh Scenarios.
Generation 3: connectivist pedagogies
The Internet’s virtual elimination of the expensive, rival nature of stored information, along
with its scale, it’s potential to connect people and a growing understanding of the significance
of socially distributed cognition (Pea, 1993; Saloman, 1993) have spawned changes in how
we perceive knowledge and the purpose of learning itself. This has led to a growth in
connectivist pedagogies of various flavours. The theory of Connectivism itself is chiefly
propounded by George Siemens (2005) and Stephen Downes (2012) but other theories and
models following a broadly connectivist pattern include, for instance, heutagogy (Hase &
Kenyon, 2007), rhizomatic learning (Cormier, 2008) and networks of practice (Wenger et al.,
2011). In connectivist-era pedagogies, learning and knowledge itself are seen as distributed
and networked, reified in shared objects that are part of the network. Learning is a process of
connection and sense-making through mediated interactions with machines and people,
directly, through shared objects, and through the formation of each learner’s networks.
Learning derives from emergent or individually specified goals, rather than through planned,
teacher-led processes. It is a result of engagement and of active creation, existing in a context
of an abundance of capacity to share, create and connect with others through networked
8
Page 9
media. Such pedagogies fit neatly into the ‘web of confidence’ quadrant of the Edinburgh
Scenarios.
Generation 0
In our initial classification, our distance education focus meant we neglected to discuss face-
to-face teaching modalities, but their consequences are profound for understanding the nature
of the next generation of e-learning pedagogies.
Face-to-face pedagogies are diverse, notwithstanding deeply embedded instructivist
cognitivist/behaviourist tendencies in institutional learning. Social constructivist patterns also
occur naturally in campus-based education through tutorials, studios, study groups and other
collaborative processes. Connectivist pedagogies, though seldom explicit in teaching, have
always been implicit on university campuses, where bars, noticeboards, cafes, smoking areas
and hallways provide the social substrate where unplanned and emergent networked learning
occurs.
However, just as distance learning pedagogies have been affected by the technologies
available, so have those of physically co-present teachers and learners. The vast majority are
based on the pragmatic assumption that a set of learners will be collocated in a classroom for
a set period of time, on a paced schedule, within the confines of a set term or semester and
will be guided by a teacher while confined there. This is an efficient way to make use of
scarce teacher time to support many learners in a physical space. Two inevitable
consequences emerge. The first is that teachers need to have much greater control than
learners: every moment in the classroom is potentially and usually guided. Second, the
chances of all learners finding every moment of every lesson to be of innate interest are very
slim. Self-determination theory suggests that lack of perceived control and insufficient/too
much challenge is antagonistic to intrinsic motivation (Kohn, 1999; Ryan & Deci, 2006).
9
Page 10
Consequently, most pedagogies of the physical classroom must be concerned with
restoring or replacing lost motivation that results from an inevitable lack of interest and a lack
of control over the process. Students may be allowed freer rein, but the key word is
‘allowed’. The most enlightened of learner-centred teachers use phrases like ‘I get my
students to…’, ‘students have to…’, ‘I make my students…’. Even in close online simulacra
of traditional classrooms such as a realtime webinar, an attendee is always in both a physical
and a virtual context uncontrolled by the teacher. Asynchronous course forums, blogs or
guided exercises afford considerably greater control. This shift of control is the fundamental
difference between online and face-to-face education. Originally a side-effect of the
educational process, the use of assessment as a means of control restores some power to
teachers, a pattern reinforced by LMSs that are based on replicating the technologies of
institutional learning (Dron, 2006).
The fundamental weakness of this replicant model becomes abundantly clear when we
look at courses where the same processes are employed without the use of threats or
punishments to drive them. MOOCs, courses that typically mirror traditional teaching
methodologies from the classroom, that are often designed with greater care and attention to
detail than their physical counterparts, that attract mature, competent and well-qualified
learners, but that offer no formal accreditation, have average completion rates of less than
10% (Jordan, 2014), regardless of pedagogy. Though some lack of commitment may be due
to reduced loss aversion (the potential loss of time, money or social face), this provides fairly
conclusive evidence that, once the pattern of control is removed from the picture, the
pedagogies that work within classrooms often have next to no value in and of themselves.
Without the assumption of control and related credentialing, it is like building a gas engine
but leaving out the gasoline. The orchestration is missing its central motif. With that in mind,
10
Page 11
we turn our attention more clearly towards the future and suggest how the next generation of
pedagogy may emerge.
THE FOURTH GENERATION OF E-LEARNING
PEDAGOGIES
We begin with the technological toolset that extends the adjacent possible. This includes,
non-exhaustively:
1. Learning analytics, adaptive hypermedia and personalization: traces of learning
activity (past and current) are captured and may then be used to either teach
teachers about their students or, more directly, guide or support students
themselves.
2. Collective technologies: traditional communication tools have been one-to-one,
one-to-many or many-to-many. Internet technologies such as collaborative filters,
tag clouds, reputation systems, network-mining algorithms and crowd-driven search
algorithms make many-to-one communication possible. The crowd, in conjunction
with intelligent aggregation of its actions, in effect becomes a teacher.
3. Deep learning and artificial intelligence: simple AI approaches have been
supporting learning for decades, from Microsoft’s Clippy to intelligent tutoring
systems. More recently, deep learning algorithms that mimic aspects of human
learning have become significantly more powerful and are beginning to provide
everything from contextual help to machine translations.
4. Disaggregated tools and services: the monolithic LMS is giving way to small,
interoperable services. The technical infrastructure ranges from older standards like
RSS and LTI (Learning Tools Interoperability) to emerging standards like
OpenBadges (enables learners to maintain personal ‘backpacks’ to collect evidence
of accredited competence from many sources), xAPI (also known as TinCan,
enables collection and aggregation of evidence of learning from any networked
source) and Caliper (a standard for capturing learning activity data used in learning
analytics and other tools, as well as service-based architectures). These standards
11
Page 12
support aggregated services as well as one-size-fits-all monoliths. The recently
announced EduCause next generation digital learning environment (NGDLE)
exemplifies this trend (www.educause.edu/library/resources/next-generation-
digital-learning-environment-ngdle). In this context, personal learning
environments (PLEs) are becoming increasingly relevant and significant, as many
have predicted for some time (Attwell, 2007). The Learning Performance and
Support System (LPSS, www.nrc-cnrc.gc.ca/eng/solutions/collaborative/lpss.html),
currently being developed by Stephen Downes and his team, is a good example of a
designed tool to provide this individually focused but highly networked toolset,
though many other more generic tools, from news aggregators like Pocket, to note
aggregators like Evernote, to Wordpress sites, already provide much of the
functionality needed.
5. Mobility and device diversity: e-learning is no longer the sole domain of personal
computers but extends to cellphones, tablets, watches, clothing and more. The Web
paradigm is being slowly eroded by the growth of apps that offer both the promise
of greater interactivity and customization, and the threat of locked-in silos
controlled by cloud-based providers. Perhaps more significantly, they open up far
more opportunities to collect data through different sensors – audio, video, location,
direction, velocity and more and, most important of all, they are with learners any
time and any place.
6. The Internet of Things and ubiquitous computing: increasingly, computation is
being integrated into our environment, and the objects we use are increasingly both
networked and smart. Because they are networked, they can interact with us in new
and complex ways, blurring the lines between the physical and the virtual.
7. Virtual and augmented reality: virtual and (especially) augmented reality has been
making inroads into the educational and training space for more than two decades.
Virtual reality has seen extensive use in the form of simulations for training as well
as in offering richer co-presence in systems like Second Life and ActiveWorlds, and
technologies like Oculus Rift and Google Cardboard are sparking renewed interest
in its potential. Augmented reality has moved out of the laboratory in the past 15
years, not just in high-profile technologies like Google Glass but in a prolific range
of mobile apps that overlay virtual information on physical spaces.
12
Page 13
8. 3D printing: a wide range of methods of creating physical objects from virtual
representations are reaching the mainstream and are as affordable as 2D printers
were less than a decade ago. As the range of materials, speed, reliability,
affordability and ease of use all increase, these further blur the lines between virtual
and physical spaces.
The first three of these trends – analytics, collectives and AI – are mainly concerned with
delegating processes to computational agents that were formerly the domain of human
beings. The fourth – disaggregation – speaks to an overlapping trend of increasing learner
control, as well as a general trend towards greater flexibility, which is increasingly enabling
alternatives to the traditional form and content of institutional learning as well as, potentially,
for learners to build their own custom learning environments. The final four trends build on
and incorporate the former but share in common an encroachment of the virtual onto the
physical. While e-learning moved the act of learning out of the classroom, it was performed
at a desk or on a laptop computer and so remained an activity that removed learners from the
flow of their lives. Learners might periodically leave their computers to engage in learning
activities elsewhere but the computer was the nexus of it all. Mobile, locative, situated and
augmented reality systems bring learning opportunities to everywhere and anywhere, making
it possible to knit an intentional learning process with the fabric of learners’ lives. Combined
with the algorithmic innovations that make such devices smarter, and the ability to assemble
and mash up pieces on demand, the patterns of teaching that assume teacher control within a
fixed space and time frame are therefore broken beyond repair. Such tools enable learning to
be embedded in practice, as part of the process of living, rather than demanding a separation
of life and education. In this way, the pedagogies suggested by these affordances bear more
resemblance to those employed by hunter-gatherer tribes and informal apprenticeships than
those used in formal teaching.
13
Page 14
We have formerly described the forthcoming generation emerging from these adjacent
possibles as the holistic generation (Dron & Anderson, 2014), in part because it will
undoubtedly incorporate parts of each of the previous generations, but also because of its
deep integration within learners’ whole lives and those of others. We are sure there is a better
term that captures its integrational quality. Though we cannot yet know in detail the
pedagogical models and theories that exploit such adjacent possibles, let alone their names,
we can extrapolate what some of the elements and characteristics of this next generation of
pedagogies and methods will be:
1. It will be focused heavily on the individual learner
Whether through adaptive learning pathways, focused feedback on individual progress, or
provisioning means for individuals to both learn and assert competence, the individual will be
at the centre. This is far from asocial, however, as social networks will often be the means
through which that individual learning is accomplished and social context will be central to
the situated nature of the activity. Every individual is part of a social network that is different
from that of every other. A social network is constituted in the connections to people we
know, so the individual is always at its centre (Wellman, 2002). This makes it very distinct
from the traditional group model of formal education, where a purpose-driven group, such as
a class or cohort, is something that can be joined, that has a name, independent of its
members, and that is rich in process and design (Dron & Anderson, 2014).
A mixed blessing in this deeper focus on the individual will lie in the trend to
personalization, typically automated through adaptive hypermedia, AI, collective intelligence
and learning analytics tools, supported by sensors to track more than just computer-based
interactions, of which smart watches and cellphones are just the beginning. There are strong
dangers that such tools, in providing a personalized path, will reduce the agency of the
14
Page 15
learner and, potentially, may impact on his or her crucial ability to learn to learn. Moreover,
as algorithms contribute more and more in guiding our learning journeys, we may see more
and more of what we want to see, and less that is neither serendipitous nor challenging
(Pariser, 2011). More positively, these tools will also make it possible, even within formal
institutional teaching, for each learner to be provided with a path that is more suited to his or
her needs. Where learning objectives are clear, this will provide great efficiencies and may
have a positive impact on the motivation of learners who will be less likely to be bored or
confused as a result. It remains to be seen whether this balances the demotivating effects
caused by the implied loss of control. We anticipate that this will become a significant area
for future research and design. Methods will be developed that are both adapted to learner
needs and that empower learners to choose between opportunities offered them (Dron, 2007)
and to be in control of the personalized student models that drive learning activity selection.
Learners will increasingly orchestrate learning themselves, through PLEs that aggregate
the flow of information from a network, organize it, support contribution to it as creators, and
support planning of learning pathways. Meanwhile, tools to aggregate evidence of learning
for individuals, rather than through institutional accreditation, will enable activities in many
different systems to be treated at an individual, rather than platform-dependent or
institutional, level.
2. It will be distributed, technically, socially and organizationally
While centralized institutions, organizations and learning management tools will continue to
play a role for the foreseeable future, both the process of learning and its accreditation will be
more distributed. This will affect both the individual’s approach to learning and the manner
of teaching. It will be rare for learning to be orchestrated through a single organization or
technology, and the role of the professional teacher in the process will be significantly
15
Page 16
disaggregated. Some may take on an orchestrating role, others may be role models, while still
others will provide an on-demand support role. The teacher will become more of a service
provider rather than a controller of the process. The centralized LMS will become far less
significant as it becomes simply one of many service providers, whether as a learning record
store or as a purveyor of content or interaction in a broader web of tools and social
opportunities.
One notable effect of this will be to change how professional learning designers, or learner
experience designers (Kop & Fournier, 2013), go about the design process. E-learning has
long relied on variants on ADDIE (analysis, design, development, implementation and
evaluation), an approach to project management inspired by the engineering models that
drove first-generation cognitivist/behaviourist pedagogies. While such approaches will not
disappear (technologies do not die), the distributed, componentized nature of the environment
will support a trend towards bricolage-based ways of building (Turkle & Papert, 1992).
Bricolage is concerned with making use of available materials and technologies. It is a means
of assembly and, as our existing technologies gain in sophistication and diversity, such
assemblies can become very rich, and be built and adapted very fast. Learners will be able to
do this too.
3. It will be crowd-driven and emergent
Learners already find inspiration, advice, help, support, guidance and other teacher-like
features distributed among the people they connect with, and such connections are expanding
exponentially. In the next generation, such assistance will increasingly not only be from
people that they know but also people that they do not. Systems like Wikipedia, Google
Search, YouTube and Q&A sites like StackOverflow are already far less network-driven than
they are set-driven (Dron & Anderson, 2014). They are driven by interest rather than
16
Page 17
connection, with cooperative rather than collaborative processes enabling people to learn
from one another. Crowd-driven support, using collective algorithms to aggregate and refine
crowd knowledge, will become more sophisticated.
At present, most approaches to using the crowd for support employ simple preference or
taste: for example, the collaborative filters of Amazon or Netflix, the PageRank of Google
Search or the ratings or ‘likes’ of countless social sites. Preference does not necessarily
indicate pedagogical value so, increasingly, we will see tools of this nature developed
specifically for learning (e.g. (Brusilovsky et al., 2004; Dron et al., 2000; Drachsler, 2009;
Graf et al., 2008; Hummel et al., 2007; Vassileva, 2004) that recommend, adapt, filter and
support with pedagogical intent. Unlike most adaptive systems so far, these will operate on an
open corpus, supported by traces left for analytics purposes and pattern matching between
learners. In keeping with deep learning approaches in general, the pedagogies such systems
use will not necessarily be explicitly programmed in, but will emerge from the data. To a
significant extent, therefore, pedagogies will not be theory-driven but data-driven, emergent
rather than designed. It is very possible that this next generation will largely be discovered
rather than invented: as we uncover patterns that appear to work, these will allow us to derive
more generalizable insights from large-scale empirical data.
4. It will be integrated, just-in-time and authentic
Without the need to enter a classroom but with the assistance of machine and human
intelligence, learning may truly be achieved any time, anywhere. Those of us who already
use cellphones to seek answers in conversations, information about places we visit, advice
from friends, even bus times or a way home that avoids the traffic, are already familiar with
the leading edge of this. Combined with smarter algorithms, crowd-sourced knowledge,
intentional support and smart sensors that know what we are doing and how we do it, the
17
Page 18
sophistication of these tools will improve and the need for separation of life and learning will
diminish, resulting in a technologically augmented return to the ways of our ancestors, in
which children learned through observing and participating in the activities of their families,
friends and others in their tribes, receiving advice, stories, correction and inspiration from
those already skilled. Such integrated approaches were formerly impossibly complex to
manage in even a moderately complex, hierarchical society, which is why schools,
universities and accompanying Generation 0 pedagogies were needed. Coming full circle,
mobile, augmented, ubiquitous, locative and always-on digital technologies return us to
integrational ways of learning for which we are naturally adapted.
5. Courses will play a less significant role
The course as an organizational structure is not built to serve learners but to enable
institutions to manage the teaching process. As technologies and pedagogies improve, this
will no longer be such a constraint. We already see a proliferation of finer-grained credits,
unpaced options, differently sized courses and alternative aggregations as a very significant
trend. Among the simpler and coarser of current alternative aggregations are the nano-
degrees offered by Udacity and the Signature Tracks provided by Coursera, which offer
small, targeted programs for developing specific ‘industry-ready’ skills, comprised of
variable-length courses. The finer the granularity, the more likely a course will be to suit
learner needs, and to be an appropriate size for the task, rather than for the timetable. Sites
offering small just-in-time learning chunks like the Khan Academy and Wikipedia, or those
offering support and answers to questions like StackOverflow or Reddit, or the ‘how-to’
videos on YouTube or Instructables, illustrate how deliberate learning activities enabled by e-
learning will become increasingly fine-grained and potentially more assemblable. This relates
to both the personal and distributed nature of next-generation pedagogies.
18
Page 19
6. Learning will be divorced from accreditation
While self-guided learning using diverse, fine-grained tools and ubiquitous access to a vast
crowd of teachers already dwarfs in quantity the formal learning that goes on in institutions, it
has, till recently, been difficult to accredit (Livingstone, 1999). The disaggregation of
learning from accreditation may be the single most important pedagogical innovation to
emerge in the forthcoming generation. Given the individualized nature of learning in the next
generation, and the reduction in significance of the traditional course, individual learners will
no longer need to be driven by accreditation needs and grades in their learning process. They
will learn what they wish to learn, when they need to learn it and, as a result of fine-grained
recording of the process and products, will naturally acquire evidence of competence and
success while doing so.
There will likely remain an ongoing need for authority to support that assertion, but it does
not need to be a single institution. One competent individual or organization might assert our
competence in one area, while an institution might support it in another, while aggregated
endorsements, records of contribution, recommendations or even likes/favourites might
provide more evidence from the crowd. It will not be necessary to receive a certificate from a
single institution and, indeed, there may be a spectrum of information about an individual’s
competence gleaned from more and less reliable sources that will provide a machine-readable
confidence profile which those that care, such as employers, may use to decide how much
they trust the assessment.
Increasingly, we will see crowd-based, data-mining approaches being used to assert
authority, along the lines of the crude methods currently used to assert reputation in Linked-
In endorsements, eBay reputations, StackExchange karma or citation indexes. Rather than
resting on institutional credibility, endorsement will come from the network, of which
19
Page 20
institutions will only be one kind of player. A mechanism like PageRank (Page et al., 1999),
which iteratively calculates relative weights for objects in a network, may make this more
reliable and enable it to scale: authority may be established by millions, rather than a few
appointed judges. This will not only help learners to find the most appropriate help, but will
also give authority to their assertions of competence.
THREATS FROM THE FUTURE
Here, we describe a handful of the more pressing issues that will continue to emerge as a
result of the next generation of pedagogies.
Open versus closed
The struggle between those that seek ways to profit from online learning and the nature of
knowledge as a non-rival social good that gains value from being spread will continue. A
combination of digital rights management (DRM) and legislative tools like copyright, patent
and trademarks will continue to support commercial providers with income. Relatedly, non-
owned rental models – typically through the provision of cloud-based services – will be
aggressively pursued by traditional and new providers.
Revenue will not only come directly. Most of the technologies we have mentioned imply
that substantial data will be collected about learners and, in many cases, could not operate at
all without doing so. Such data may be owned neither by the student nor by a trusted
institution. Already, complex moral and legal issues surround, say, requiring student
engagement on commercial social media sites or publisher-owned e-textbook systems. In
many cases, the traces left by learners through standards-based aggregation mechanisms will
be more commercially valuable than the services they purchase, particularly in advertising
revenue. In search of ownership of such data, large publishers and other commercial
organizations will compete with traditional institutions for parts of the learning market, likely
20
Page 21
in partnership with tools providers like Blackboard, Canvas or Desire2Learn. Rather than
paying one provider, learners may need or desire to spread learning costs across multiple
providers. Perhaps, micro-payment systems like BitCoin may enable individual experts to
offer on-demand services that are part of this trend. It is likely that aggregation and
integration services intended to simplify such issues will become significant: universities and
colleges may well play such a role. How or whether such efforts will compete with copious
free support provided by individuals remains to be seen. As now, support for accreditation
may be the most valued service, whether distributed or not.
It is equally certain that open educational resources will play an ever-more important role
in both formal and informal learning. When learning resources are open, they may be
modified and re-shared, following a Lamarckian evolutionary dynamic. Prolific replication
with variation and competition for limited attention means the best will thrive while the worst
will decline. This means they will continue to improve, diversify and spread. The benefits of
open educational resources already greatly exceed those of traditional closed resources
(Wiley & Hilton III, 2009). This is exactly the dynamic that has resulted in the almost total
dominance of Wikipedia over traditional encyclopaedias and its emergence into information
realms never penetrated by paper-based encyclopaedias. We may see more public funding
diverted to such efforts as their role as a social good begins to be seen on a par with
traditional universities and public libraries, and as those older institutions continue to
diminish in relevance.
The loss of mind, the loss of soul
Perhaps the most vague and yet the most troubling of threats to the future of e-learning comes
from the backlash from prominent authors (e.g. Brabazon, 2007; Carr, 2011; Greenfield,
2004; Harris, 2014; Keen, 2007; Turkle, 2011) who bemoan the very changes that we have so
21
Page 22
far celebrated. They argue that we are becoming less focused, drawing shallow content from
diverse sources, our critical skills are being diminished, our social skills becoming
caricatures, and our connection with others diminished in quality while quantity increases.
All technologies change us, especially those affecting something as fundamental as how we
learn, and not all of those changes will be positive.
To handle such big issues in a small chapter is over-ambitious. Instead, we focus on a
narrow but illustrative example. Some of the more disruptive effects of e-learning have been
to demolish the foundations of traditional teaching assignments and examinations, making
plagiarism, real-time examination answer-sharing and contract cheating even more ubiquitous
than before. A whole industry has grown up around contract cheating, with many sites
dedicated to providing custom-written assignments for a fee. This has led to an unwinnable
arms war. Over two-thirds of American college students admit to cheating within the last
year (McCabe et al., 2012) and, in some countries, the rates are considerably higher.
In the first place, this makes a nonsense of assignments that rely on easily reproducible
facts or simply discovered writings. In the second place, it challenges the role of universities,
colleges and schools as accreditation agencies and is one of the main reasons that we foresee
a bleak future for that role. Traditional accreditation methods make no sense in a context of
abundance, where the best of traditional educational outputs is often easily outclassed by the
worst of Internet inputs. The focus of both courses and students on accreditation is a
consequence of Generation 0 pedagogies that are broken by the distributed, personal,
ubiquitous, situated, data-intense world of the coming generation. In microcosm, this battle
is an analogue of the changes that are radically changing how we learn and how what we
learn is accredited – both a symptom of the change and a catalyst for it.
22
Page 23
CONCLUSION: PEERING INTO THE FURTHER FUTURE
FROM THE DISTANT PAST
The known
Some of the future of e-learning is very knowable: it will be like the past. Technological
revolutions seldom if ever fully replace their forebears and almost always incorporate parts of
them. The same tools and methods that we use now will continue to exist, just as remnants of
archaic technologies of traditional education, libraries of books, lectures, classes and
medieval trappings of academia, still persist, not just physically but in our LMSs. As always,
the old will accompany, be integrated with and aggregated with the new.
At the beginning of the 19th century, Pestalozzi influentially sought techniques to teach
that were largely independent of teachers:
I believe it is not possible for common popular instruction to advance a step, so long
as formulas of instruction are not found which make the teacher … merely the
mechanical tool of a method, the result of which springs from the nature of the
formulas and not from the skill of the man who uses it. (Pestalozzi, 1894, p. 41)
Traditional schools and universities are designed as machines for learning that deliberately
harden the process. The use of timetables, curricula, tiered lecture theatres, corridors and
classrooms are all designed to reduce possible choices for the teacher as well as the learner.
Technologies can equally soften and support creativity, however. For instance, the invention
of the blackboard augmented the teacher’s capacity to explain and library indexing systems
brought powerful tools for exploration.
The capacity of electronic devices to be universal tools, media and environments affords
both softening and hardening. From early Skinnerian behaviourist teaching machines to the
latest learning analytics tools, hard, prescriptive tools that play one or more roles of the
23
Page 24
teacher are ubiquitous. Equally common are softer holistic tools, from electronic whiteboards
to webinars, which enable teachers (including learners) to do more, in new and different
ways. This tension between softening, creative opportunities and hardening, efficiency-
building approaches to learning will undoubtedly continue to play out in the future of e-
learning.
The unknowable
In a boundaried system, be it a school, a university or an individual learner, if changes in the
ecosystem of which it is a part occur faster than the boundaried system itself, its chances of
survival are slim. Those boundaries are changing very fast indeed, as the relentless logic of
the adjacent possible predicts they must. Some shifts may be profound. For instance, it is far
from ridiculous to imagine a future in which we might, for example, take a pill to learn
Spanish or learn to read (Negroponte, 2014). There are enormous ethical concerns with doing
so – rewiring brains to adopt or influence a particular political, religious or social perspective
is an extremely dangerous idea, quite apart from the potential for viruses and deliberate
manipulation, and we predict enormous resistance to such technologies. Nonetheless, the
eagerness with which many of us enhance our brains with chemical, electrical or magnetic
stimulation suggests a willing market.
Such near-adjacent possibles bring into question what kinds of pedagogy may emerge as a
result. The focus of learning would be far more on doing and connecting ideas, on creation
and creativity, on building learning networks, on working with other people and on
developing approaches to the construction of ideas and objects rather than memory and skills.
With less need to concern ourselves with the mechanics of learning facts and basic skills,
learning would become more concerned with application, and thus with aesthetic and moral
values. Though the notion of an educational institution as we know it, including both its
24
Page 25
teaching and accreditation role, might seem profoundly archaic in such circumstances, it may
be exactly what is needed to bring stability and value to this brave new world. Perhaps that
continuity and integration will be institutional education’s lasting gift to society when its
traditional teaching role of skill and knowledge transfer is gone. Perhaps that is what it should
have always been.
REFERENCES
Allen, I. E., & Seaman, J. (2013). Changing course: Ten years of tracking online education in
the United States. Retrieved from
www.onlinelearningsurvey.com/reports/changingcourse.pdf
Anderson, T., & Dron, J. (2011). Three generations of distance education pedagogy.
International Review of Research in Open and Distance Learning, 12(3). Retrieved from
www.irrodl.org/index.php/irrodl/article/view/890
Arthur, W. B. (2009). The nature of technology: What it is and how it evolves (Kindle edn).
New York: Free Press.
Attwell, G. (2007). Personal learning environments: The future of eLearning? eLearning
Papers, 2(1), 1–7.
Bell, M., & Stewart, C. (2004). The Edinburgh Scenarios. Retrieved from
www.internettime.com/lcmt/archives/001121.html
Brabazon, T. (2007). The University of Google: Education in a (post) information age.
Aldershot: Ashgate.
Brooks, D. (2012). The campus tsunami. The New York Times, 4, A29.
25
Sharon Cawood, 10/05/15,
Where missing, please provide accessed dates for web pages.
Page 26
Brusilovsky, P., Chavan, G., & Farzan, R. (2004). Social adaptive navigation support for
open corpus electronic textbooks. Proceedings from AH 2004, Eindhoven.
Carr, N. G. (2011). The shallows: How the Internet is changing the way we think, read and
remember. London: Atlantic Books. Retrieved from https://books.google.ca/books?id=9ta-
uQAACAAJ
Cormier, D. (2008). Rhizomatic education: Community as curriculum. Innovate: Journal of
Online Education, 4(5), 2.
Cross, J. (2008). Origins of ‘eLearning’. Retrieved from
www.internettime.com/2008/11/origins-of-elearning/
Dewey, J. (1916). Democracy and education. New York: Macmillan. Retrieved 21 May 2001
from www.ilt.columbia.edu/projects/digitexts/dewey/d_e/contents.html
Downes, S. (2012). Connectivism and connective knowledge: Essays on meaning and
learning networks. Stephen Downes Web, Commentary, 19 May. Retrieved 5 October 2015
from http://www.downes.ca/files/books/Connective_Knowledge-19May2012.pdf
Drachsler, H. (2009). Navigation support for learners in informal learning networks. PhD
thesis, Open Universiteit Nederland, Heerlen.
Dron, J. (2006). Any color you like, as long as it’s Blackboard®. Proceedings from E-Learn,
Honolulu, HI, 13–17 October.
Dron, J. (2007). Control and constraint in e-learning: Choosing when to choose. Hershey,
PA: Idea Group International.
26
Page 27
Dron, J. (2009). Pedagogies as educational technologies. World Conference on E-Learning in
Corporate, Government, Healthcare, and Higher Education 2009, pp. 2120–2127. Retrieved
from www.editlib.org/p/32777
Dron, J., & Anderson, T. (2014). Teaching crowds: Social media and distance learning.
Edmonton, AB: Athabasca University Press. Retrieved from http://teachingcrowds.ca
Dron, J., Mitchell, R., Siviter, P., & Boyne, C. (2000). CoFIND: An experiment in n-
dimensional collaborative filtering. Journal of Network and Computer Applications, 23, 131–
142.
Graf, S., Kinshuk, & Liu, T.-C. (2008). Identifying learning styles in learning management
systems by using indications from students’ behaviour. Proceedings of Advanced Learning
Technologies, pp. 482–486. IEEE: ICALT’08.
Greenfield, S. (2004). Tomorrow’s people: How 21st-century technology is changing the way
we think and feel. London: Penguin.
Harris, M. (2014). The end of absence: Reclaiming what we’ve lost in a world of constant
connection. HarperCollins Canada. Retrieved from https://books.google.ca/books?
id=4DJ0AwAAQBAJ
Hase, S., & Kenyon, C. (2007). Heutagogy: A child of complexity theory. Complicity: An
International Journal of Complexity and Education, 4(1), 111–118. Retrieved March 2008
from
www.complexityandeducation.ualberta.ca/COMPLICITY4/documents/Complicity_41k_Has
eKenyon.pdf
27
Page 28
Hummel, H. G. K., Van Den Berg, B., Berlanga, A. J., Drachsler, H., Janssen, J., Nadolski,
R., & Koper, R. (2007). Combining social-based and information-based approaches for
personalised recommendation on sequencing learning activities. International Journal of
Learning Technology, 3(2), 152–168. Retrieved from
http://inderscience.metapress.com/content/A1302V75036X4H02
Jordan, K. (2014). Initial trends in enrolment and completion of massive open online courses.
The International Review of Research in Open and Distributed Learning, 15(1).
Kauffman, S. (2000). Investigations (Kindle edn). New York: Oxford University Press.
Keen, A. (2007). The cult of the amateur. London: Nicholas Brealey Publishing.
Kelly, K. (2010). What technology wants (Kindle edn). New York: Viking.
Kohn, A. (1999). Punished by rewards: The trouble with gold stars, incentive plans, A’s,
praise, and other bribes (Kindle edn). New York: Mariner Books.
Kop, R., & Fournier, H. (2013). Developing a framework for research on personal learning
environments, eLearning Papers. Retrieved from
www.openeducationeuropa.eu/en/elearning_papers
Livingstone, D. W. (1999). Exploring the icebergs of adult learning: Findings of the first
Canadian survey of informal learning practices. Canadian Journal for the Study of Adult
Education, 13(2), 49–72. Retrieved January 2006 from
www.oise.utoronto.ca/depts/sese/csew/nall/res/cjsaem.pdf
McCabe, D. L., Butterfield, K. D., & Treviño, L. K. (2012). Cheating in COLLEGE: Why
students do it and what educators can do about it. Baltimore, MD: Johns Hopkins University
Press. Retrieved from https://books.google.ca/books?id=DfBLxVIq5IAC
28
Page 29
NCES. (2013). Digest of Education Statistics. Retrieved from
https://nces.ed.gov/programs/digest/d13/tables/dt13_206.10.asp?current=yes
Neal, L. (2007). Not all the world’s a stage: Hi-def video will only revolutionize e-learning
when students get comfortable on camera. eLearn, 8, 1.
Negroponte, N. (2014). A 30-year history of the future [Video]. TED.
Page, L., Brin, S., Motwani, R., & Winograd, T. (1999). The PageRank citation ranking:
Bringing order to the web. Retrieved from http://ilpubs.stanford.edu:8090/422/
Pariser, E. (2011). The filter bubble: What the Internet is hiding from you (Kindle edn). New
York: Penguin.
Pea, R. (1993). Practices of distributed intelligence and designs for education. In G. Saloman
(ed.), Distributed cognitions: Psychological and educational considerations (pp. 47–87).
Cambridge: Cambridge University Press.
Pestalozzi, J. H. (1894). How Gertrude teaches her children (L. T. Hooland & F. C. Turner,
trans.). Syracuse, NY: Swan Sonnenshchein & Co.
Ritzer, G. (2013). MOOCs and the McDonaldization of education. Retrieved from
https://georgeritzer.wordpress.com/2013/01/12/moocs-and-the-mcdonaldization-of-
education/
Rosenberg, S. (1992). Virtual reality check digital daydreams, cyberspace nightmares. San
Francisco Examiner, 19 April, p. C1.
29
Page 30
Ryan, R. M., & Deci, E. L. (2006). Self-regulation and the problem of human autonomy:
Does psychology need choice, self-determination, and will? Journal of Personality, 74(6),
1557–1585.
Saloman, G. (1993). Distributed cognitions: Psychological and educational considerations.
Cambridge: Cambridge University Press.
Siemens, G. (2005). Connectivism: A learning theory for the digital age. International
Journal of Instructional Technology and Distance Learning, 2(1). Retrieved from
www.itdl.org/journal/jan_05/article01.htm
Taylor, P., Parker, K., Lenhart, A., & Patten, E. (2011). The digital revolution and higher
education: College presidents, public differ on value of online learning. Retrieved from
www.pewinternet.org/files/old-media//Files/Reports/2011/PIP-Online-Learning.pdf
Turkle, S. (2011). Alone together. New York: Basic Books.
Turkle, S., & Papert, S. (1992). Epistemological pluralism and the revaluation of the
concrete. Journal of Mathematical Behavior, 11(1), 3–33. Retrieved from
http://papert.org/articles/EpistemologicalPluralism.html
Vassileva, J. (2004). Harnessing P2P power in the classroom. Proceedings from ITS 2004,
Maceio, Brazil.
Vermunt, J. D., & Verloop, N. (1999). Congruence and friction between learning and
teaching. Learning and Instruction, 9(3), 257–280.
Vygotsky, L. S. (1978). Mind and society: The development of higher psychological
processes. Cambridge, MA: Harvard University Press.
30
Page 31
Wellman, B. (2002). Little boxes, glocalization and networked individualism. In M. Tanabe,
P. van den Besselaar, & T. Ishida (eds), Digital Cities II: Second Kyoto workshop – revised
papers. Springer. Retrieved from
http://homes.chass.utoronto.ca/~wellman/publications/littleboxes/littlebox.pdf
Wenger, E., Trayner, B., & Laat, M. D. (2011). Promoting and assessing value creation in
communities and networks: A conceptual framework. Ruud de Moor Centrum/Open
Universiteit. Retrieved 5 October 2015 from
http://wenger-trayner.com/documents/Wenger_Trayner_DeLaat_Value_creation.pdf
Wiley, D., & Hilton III, J. (2009). Openness, dynamic specialization, and the disaggregated
future of higher education. The International Review of Research in Open and Distributed
Learning, 10(5). Retrieved from www.irrodl.org/index.php/irrodl/article/view/768
31