1.
Introduction
This paper is based on work in designing and building
learning materials and environments in a variety of contexts:
-
Knowledge bases for
community and vocational training in developing countries in the Middle East
and South Asia;
-
Learning materials for
schools in Southern Africa;
-
e-learning for open and
distance learning courses in the UK;
-
Research on strategy and
evaluation tools for e-learning, expert systems and knowledge management.
It shares with many
others the frustration at systems that “do” e-learning, but for the most part
do it only marginally differently from what we did 10 or 20 years ago using
other quite adequate media, and in the process generating huge amounts of
expensive digital traffic, data mountains/lakes/landfill sites, and Public
Relations, but not much digital intelligence. I will try to tease out some
lessons and issues on how we might do things differently if we started from
different design parameters.
2.
Distributed Learning
There is no doubt that
digital or e-learning, or e-enabled learning has made a difference to our
lives, and that it is largely based on digital information and communication
technologies. However, as with all other innovations, ICT will go through
cycles of development, hype, overshoot, disillusionment, shake out, and
consolidated growth. We are probably currently at the disillusionment and
shake out point, and will start to see more tentative, consolidated growth
from now on.
But the crux of
managing this turning point [for ICT like any other technology] is to realise
that the key issues are not technical or technological. Once the “wow” factor
has soured and faded, we need to focus on how we want to use it, sensibly, and
why. There are some crosscutting trends that might point us in the right
direction.
In overall terms, what
has developed more than anything else is distributed learning. The fact
that it is powerfully, cheaply, and easily (?) distributed through integrated
digital media is important. But the levels of digital integration have only
just started, and it is the result – distributed learning – rather than
the mechanism – digitalisation, that we should focus on.
There is and will
always be proprietary information and knowledge. But most of the old exclusion
techniques and cycles are changing or fading: you can no longer be excluded by
age – younger and younger people are getting hold of all the information they
need and want, and quicker. Likewise distance, nationality, religion,
wealth/parentage, “outlaw” status, libraries, academies, country, cost,
sequence and ritual, and even access to technology are all being penetrated.
Learners are better
sited and sighted – they are quite simply in a better position to see what is
going on around them, interact with it and with people who are involved with
it, and learn from that. They are also better “cited” as they can learn from
and refer to a much wider range of people and texts, more quickly and
effectively. Some of the ideas that I will use in writing this paper, for
instance, I got from a casual reading of various Blogs/Klogs a few hours ago.
So what matters is that
learning is distributed – the medium does not matter. The digital media are
well established. We have arrived at quite a different media and
communications platform – so we no longer need to be fascinated by the latest
medium or gadget, and should get on with the job of using whatever we have –
blended media and blended learning: blended-distributed-learning.
Following from this
trend, i.e. comprehensively distributed learning, is the second trend, namely
that not only is the blend important, but none of us are at the
point of control – there is no single point of control – there are
many, competing points at which learning and learning resources gets
initiated, stimulated, certified, commodified, patented, copyrighted, shared
illegally or legally, accredited, and all of them are changing.
Learning and learning
resources are [and will increasingly be] distributed to an unprecedented
extent. All bets on the old walls and fences around learning are off. This is,
surely, positive. If so, we need to go with the flow, however much our
institutional, personal, and patronage practices get disrupted. As I said
above, there will always be proprietary and public learning resources and
opportunities – and there will therefore always be business and work in
enhancing learning. And conversely, there will always be interesting
developments in sharing knowledge – developing the “new commons” of the
knowledge society.
3.
Just-in-Context Knowledge
Snowden (2002:3) citing
Stacey (2001) talks of the paradoxical nature of knowledge, which is both a
thing and a flow or a process, and he emphasises that we
have to see it as both – not the one or the other. He also cites some
key heuristics: “knowledge can only be volunteered” and “we only know what we
know when we want to know it’, and he emphasises the value of narrative.
Knowledge is thus
embedded in relationships and context, and little of it is amenable to
commodification and categorisation in a database, no matter how sophisticated.
Most of it is situated in the spaces of the relationships between human
beings. And narratives come back to take up their place alongside algorithms,
just as they did in Athenian discourse.
This is similar to the
notion that knowledge is essentially strategic, and information is
essentially procedural (Williams 2001). Knowledge subsumes and includes
content as well as complex procedural algorithms, but it is more akin to
intelligence than information
It operates within a context, and is implemented or used by particular people
in particular positions and contexts. To paraphrase, knowledge is a synthesis
of the how and the why things get done, whereas information
stops at the how. Knowledge is paradoxically more contextualised, and
therefore less abstract than information, even though it operates at a
meta-information level.
Knowledge is embedded –
it is what I would call “just-in-context”. This means that it is specific to
time, place, sequence and timing, and position and relationships – within
discourse communities/communities of practice, and personal relationships of
trust and confidence. Your ability to exercise what
you know is partly dependent on the fact that I know what you know, that you
know that I know that, and that I trust you to use it appropriately.
It follows that
knowledge cannot be abstracted from context – physical or social. Snowden (op
cit: p3) says that to manage knowledge “we need to focus more on context and
narrative, than on content”. In relation to e-learning, or distributed
learning, what is important is that we don’t restrict learning to abstracted
procedural information, and call that knowledge. Learners need to develop
their own knowledge, through a process of learning that will include
procedural information, but which they must relate to various contexts –
contexts in which it is generated, learnt, used, and in which they can use it.
Algorithms can only be understood and used within narratives.
4.
e-learning
Where have we got to in
e-learning? There is undoubtedly much good and innovative practice. But there
are also many cases in which only the technical transition has been made, and
not much more. Lots of “e-learning” would more appropriately be called
“e-copying/photostatting”, “bookware”, e-distribution, e-searching, and
perhaps e-publishing. But this is not much more than increased efficiency on
the supply side. It does not necessarily impact on effectiveness in satisfying
learner demands, and in line with the trends in down-loading and
externalising costs from the public sector to the public, it might even
decrease effectiveness in some cases. Some members of the public can’t afford
the externalised costs, and are unimpressed that the internal “costs” have
been reduced to help reduce taxes and fiscal deficits.
It is commonplace now
that we are suffering from information overload, as well as email-induced
communication and interaction overload. From a systems design point of view,
we need to determine the learning parameters of digitising distributed
learning. If the aim of learning is to explore your surroundings to accumulate
useful knowledge, what is it that contributes to learning and knowledge – over
and above faster copying, searching and distribution? This might be called
e-linking, e-relationships and networks, and e-enhanced strategy.
5.
Designing Distributed Learning
The basic design
parameters are:
Linking, analysing
and synthesising at a conceptual level helps to facilitate learning, and to
capture and manage the knowledge that results from learning. Relationships
require interaction, including both intellectual and personal relationships.
And strategy requires a synthesis of information about procedures and context,
and the experience and knowledge against which to measure and evaluate them.
If we know that
knowledge includes content, context, and relationships; that learning requires
exploration, and links at the conceptual level, as well as personal and
intellectual interaction, and the ability to manage information about
procedures and contexts against the template of experience, then we have the
beginnings of a framework for the design of distributed learning.
Exploring includes
finding out what is out there just as much as putting out your “feelers” –
physically and intellectually, to see what happens. So in a digitised world of
distributed learning, we should use all the digital and analogue media,
particularly as they become cheaper and more user friendly – email, websites,
weblogs, digital video and webcams, digital photography, and so on.
Links at the conceptual
level need to be more than just linear-embedded “threads”, elegantly “woven”
by e-moderators. However interactive these are, and they do enable valuable
virtual communities or virtual “classes” to flourish, they are no more than
stacked lists, or what I call “stringed-bead discourse” – a number of
“beads”, each one of which is only linked to the one before and after it, with
little or no relation to any of the other “beads”, apart from some social and
stylistic “aesthetics”. The same applies to the electronic “filing systems”
that are available in word-processing packages (albeit with some primitive
hyperlinks available).
What is needed is more
than a linear architecture – and there are two-dimensional graphic options in
some of the e-mind-map packages available, but these are hardly mainstream in
VLEs.
What is needed is a
two-dimensional plane on which learners and teachers can explore, elaborate,
rearrange and restructure, link and question, the relations between concepts
and contexts, with dynamic granularity and navigation [which just means that
you need to be able to navigate “free-hand” and zoom in on any point of the
plane just as you would when using a digital photography cropping facility]. A
facility to establish icons alongside and/or related to objects on this plane
is also needed. Behind this (in XML format) there needs to be a data base,
linked to the metadata on this plane. And this needs to be available in a
collaborative workspace format.
This could be developed
further, based on some of the available software. It would start to deliver
some of the dynamic metadata links that are the basis for any non-digital,
non-technical learning and knowledge, in those rather sophisticated “neural
network processors” called humans.
5.1
Just-in-Time Informal-Formal Knowledge
Snowden (op cit)
outlines a framework for facilitating just-in-time knowledge. His paper is
very useful, as he details a framework for analysing four domains of
knowledge, each with different management implications, and then relates
just-in-time transfer to those domains. I wont repeat the details – they are
quite extensive, and there is no need to try to summarise them here.
5.2
Decision Making
Given that the aim of
learning is useful knowledge, two things are necessary for learners to manage
their own knowledge as they learn. They need to be able to capture and
manipulate the links that constitute learning and knowledge on an appropriate
high granularity graphics package. Second, they need to be able to capture the
knowledge, and access it efficiently, for use. Which means they need to be
able to search through the information available quickly and effectively.
That in turn means that
the knowledge that is commonly held should reflect the decision-making
processes that someone using that information and knowledge would
follow. Given that the conceptual map is metadata, it would not inherently be
a problem for there to be a variety of different maps, each for a different
type of user. The users in turn could customise their maps, and the links from
them to the database, in line with the specifics of their context, as they
change from time to time. So the software would be required to include the
necessary dynamic editing facilities.
6.
Sharing Knowledge
I said, above, that
there will always be proprietary and public learning resources and
opportunities – and there will therefore always be business and work in
enhancing learning. And conversely, there will always be interesting
developments in sharing knowledge – developing the “new commons” of the
knowledge society.
One of the paradoxes in
distributed learning is that we expect people working in this sector
(certainly the public sector part of it) to share learning resources within a
competitive and increasingly commercialised market, in which their own jobs
and livelihoods are neither protected nor guaranteed.
If it is true that
distributed learning no longer has a single point of control, and that
distributed learning increasingly occurs between a myriad of different points
– some human and some digital – then it would make sense to optimise the
potential for sharing information and knowledge, if possible.
Most commentators (see
David Gurteen’s website for examples) seem to agree these days that only “a
limited amount of knowledge can be fully separated from its owners and
transferred to the best practices domain” (Snowden op cit p13). And the
increasingly “marketised” and commodified environment would not seem to make
sharing possible.
Both teachers and
learners face a paradox: how can they be expected to share information and
resources in a competitive, commercial market? Action research in designing
and implementing a knowledge base for vocational training for the community
sector in the Middle East has started to put a working model together. This
knowledge base design could be used to solve this paradox, and as the basis
for national and international networking and sharing in schools and in the
vocational training and community sectors.
Schools and community
sector programmes are expected to “share resources for the common good”. But
they are also expected to maintain their own (competitive, dare one say it?)
value added, and they have to keep an eye on their own careers, in a job
market where there are few certainties and even fewer guarantees. Researchers
based in Reading (Williams and Carmichael) worked with a training centre in
the Middle East to identify their training and learning problems, and then to
design and implement the software for a knowledge base appropriate to their
needs.
6.1
The training centre wanted to:
- Use the vast amount of learning
material available internationally, but adapt it to local context and
requirements.
- Share resources as far as possible,
but maintain a competitive advantage.
- Add to available resources, and
expand its business, while sharing (both ways) with outsourced trainers and
external institutions.
- Build up a resource which could be
used for specific, narrowly defined training requirements, but could also be
used flexibly for different needs, clients, and contexts without starting
from scratch each time. The output also needed to be available in Arabic and
English texts and scripts, and displayed in various formats.
- Enable trainers and training
managers to explore and learn from related research, evaluation, theory, and
other institutions.
6.2
Stages in the process:
- The core users were narrowly
defined: as trainers and training managers (and less so, administrators).
The elements of the system had to be checked to see if they would fit into
the organisation’s workflow on a daily basis.
- The key training and learning
elements had to be defined, and if possible separated into different layers
of resources, that could be combined, revised, and added to, in separate
operations. These were “learning objects” such as pictures, texts, graphics,
exercises etc, “lesson plans” which combine learning objects, and “courses”
which combine lesson plans. Lesson plans have related “learning outcomes”
and assessment frameworks linked to them.
- The outputs needed to be defined,
and the software mechanisms created to link the resources into lessons and
courses, and provide the outputs.
- Further links need to be put in
place to related research, theory, evaluation, and institutions and people.
6.3
Findings
- Consistent institutional backing
and commitment is critical to success or failure.
- Open source software development
can provide the flexibility to combine languages, scripts, and learning
objects and lesson plans, effectively, without the expense and rigidity of
proprietary packages.
- Most of the “learning objects” can
be shared in the pubic domain.
- Most of the “lesson plans” that use
these learning objects can be kept in the private/competitive domain where
necessary. This distinction between mostly public domain learning objects,
and mostly private domain lesson plans and courses enables substantial
sharing, and flexible “tagging” of resources to indicate how and where they
may or may not be shared.
- Substantially increased flexibility
can be achieved in combining elements in the resource base and “tagging”
them if XML (rather than HTML) is used – and this does not affect the
ability to access the resources via any Internet browser.
7.
Conclusion
I have sketched some of
the basic elements of a framework for an integrated approach to the design of
distributed learning and just-in-context knowledge management. The learning
and knowledge issues have, hopefully, been separated out from the technical
issues and the technology, and then reinserted into a debate about how the
learning design parameters could be realised and developed with current
capability, and how the potential for distributed learning to benefit from
shared resources could be managed within an increasingly commodified world.
|
