DICK MOORE ON AGILE LEARNING, AGILE SOFTWARE DEVELOPMENT AND THE MOBILE INTERNET

Dick Moore
Dick Moore

Dick Moore was, until a week or two ago, Director of Technology atUfi/learndirect — I've known him since his days at Doncaster and Sheffield Colleges, and then at a Santa Monica dotcom. We started talking about Agile Learning shortly after my last post about it, and I quickly felt there were enough interesting ideas in the conversation to make it worth developing and sharing. So I suggested doing an interview with Dick, and sent him a few questions to think about…
A few days ago, I received a comprehensive document with Dick's thoughtful replies. Hence what follows is as much a 'guest post' as an interview, though we did have a follow-up chat. Dick has a deep understanding of tech infrastructure and methods, so the first section is his take on the possible mappings between Agile Software Development and Agile Learning. The later sections weave together Dick's answers to my questions with some additional material from our chat.
Dick's blog is at ToolsAndTaxonomy.com and you can mail him at dmoore [at] MooreAnswers dot co dot uk. [Update, 15 July 2010: Dick has now posted his version of this interview, so have a look and check for comments there as well.
I'm interested in doing a series of interviews like this, with people who have different contexts for, and angles on, Agile Learning. If you'd like to be interviewed, please get in touch.

Agile Learning and Agile Software Development

What do we mean by Agile Learning? In software development, the 'agile' movement was as a reaction against large scale development projects governed by a monolithic organisational standard perceived to be overly bureaucratic, costly and slow for what is often small scale development. Not all development is suitable for such an approach in much the same way that not all learning and assessment could be considered suitable for an agile approach (though there may be elements within large learning programmes that might benefit from agile methods to better reflect real world situations).

The agile development manifesto defines some core values:
  • Individuals and interactions v processes and tools
  • Working software over comprehensive documentation
  • Customer collaboration over contract negotiation
  • Responding to change over following a plan
And here is my [Dick Moore's] take on how the principles of Agile Learning might relate to those of agile development:
Agile Development
Agile Learning
Customer satisfaction by rapid, continuous delivery of useful software and systems
Learner satisfaction by rapid attainment of learning concepts that can be applied
Working software and systems delivered frequently (weeks rather than months and years)
Attainment of new models of understanding and assessment building upon each other in short durations (months)
Working software and client satisfaction are the measure of progress
The ability to apply and contextualise learning with clear signs of progress and development
Late changes are welcomed rather than rejected out of hand
The ability to change particular learning goals as understanding or issues arise
Close, daily cooperation between clients and developers
Close relationship between educators and learners (often with blurred roles)
Face to face conversation is the best form of communication requiring co-location)
Regular communication (daily/weekly) mixing synchronous and asynchronous communication as a key feature, and augmented via technology
Projects are built around motivated individuals who are trusted
There has to be shared vision and common goal for the learning activity
Continuous attention to technical excellence and good design
Having well defined goals and structure and of "high quality"
Simplicity
Clear objectives, though still open to change
Self organising teams of 5-9 to facilitate development
As per Bloom's "two sigma problem," mastery learning can be applied in small groups, with strong communication
Regular adaption to changing circumstances
OK to change learning goal or aim mid-session providing its agreed
There is no single tool set rather a collection of tools and processes that support agile development
No one method or way of being an agile learner or supporting Agile Learning, but they require a goal and some organisation
David Jennings (DJ): Can you say a bit more about the conditions under which learning goals can change, and where Bloom's mastery learning fits into the picture?
Dick Moore (DM): We can build business models, and possibly learning experiences, in an evolutionary fashion. You set out with a clear goal, and iterate as you go to make sure that you're still on track. In the process you learn something; quite possibly that this is the wrong way to do it, or that you're asking the wrong question! SeeGetting to Plan B for more on the perspective that encourages this evolutionary thinking.Having clear goals, which are easily tested so you know when you've reached them (or not), does not mean that you can't change the goals — as long as you do so with consent of the interested parties, having demonstrated to their satisfaction that the goals were wrong.
In the 1980s, Benjamin Bloom wrote a paper on the significant improvements in achievement made possible by alternative forms of instruction. This points to the fact that education has yet to mirror the kinds of innovation that we've seen in other sectors — retail, healthcare, manufacturing — in recent decades. Teaching perhaps needs to become more content-centric, rather than teacher- or classroom-centric — though this is not to say that I want to do away with teachers and mentors and suchlike. It's just about putting the goals at the centre of the learning, and not the classroom experience.

You said that Agile Learning already exists — how, where and what's making it spread?

The need to rapidly acquire new skills and knowledge, combined with the knowledge engine that is the internet, promotes self-directed learning — be it formal, informal or recreational. Access to how-to videos via YouTube, and the previous whole set of how-to documentation that underpinned the open source Linux development platform have shown that recipes, plus a learning goal, can form the basis of significant learning and development programmes.

You distinguish between shallow and deep learning. To what extent do you think Agile Learning is destined to focus on shallow experiences?

There is a view that says that every Google query is a piece of shallow Agile Learning (after all why would you post a query if you already knew the answer (other than to prove to someone else that you know the answer). There is a goal (a question requiring an answer), a tool (search engine) and a need for some analysis of content (the results returned by the engine) with synthesis or application.
This is both agile and clearly shallow learning, though effective use of search engines may require a range of core agile skills (formulation of queries, how to judge the providence and veracity of what is returned, and basic technology skills that so many take for granted but are not yet universal.
If we accept the above as an example of Agile Learning then it's clear that most Agile Learning will be shallow in nature. However this does not mean that with a slightly larger tool set, a more complex learning goal and group of motivated individuals these techniques could not result in significantly deeper learning (there are many similarities here with the Oxbridge seminar model).
DJ: What do you think are the "core agile skills"? Do you think they're similar to the literacies of self-organised learners that I blogged previously, and what proportion of people do you think have these skills?
DM: It's not whether you have the skills or you don't; it's a degree of competence. Anyone who uses Google will have them to a greater or lesser extent. Most people will know they have to discriminate the reliability of information they find on the web, though they won't necessarily know all the different ways you might do this. Even sophisticated users can be misled by sources bent on deception.

How important is it to have a plan for learning, and what kind of planning do you think would be right for Agile Learning?

This has to be one of the core features of Agile Learning, having a clear goal that one or more people can focus around in short iterations and having a way of measuring this provides an end point, a measurable output and a sense of achievement. Having clear criteria that define the end of a learning iteration can only be a good thing.
Examples might be along the lines of
  • after this iteration I will be able to…
  • after this iteration I can demonstrate…
  • after this iteration I can explain and show the relationships between…

What are the challenges with assessing Agile Learning?

The challenges with Agile Learning are very much the same as with any other form of learning. If certification is required, then there has to be some sort of strong and rigorous assessment that underpins the knowledge and practical skills being taught.
Issues such as personation, plagiarism, weak testing regimes, and corruption all apply. However in Agile Learning we have some advantages in that clear goals are set, intervals are defined and typically short and there may be an end result delivered via a group allowing the group to self-assess (something that should be encouraged). Moderation may of course be needed.
For more formal qualifications then formal assessment might be considered. The UK Driving Test with its theory and practical test might be considered a good example of Agile Learning. I have just completed the RYA Day Skipper certificate again containing a theoretical and practical elements, delivered in several short sessions with clear goals. Both are great examples of Agile Learning and both high stakes when you consider the implications of poor theoretical or practical skills.
DJ: Those are interesting examples, because they're not the kinds that I would first think about as cases of Agile Learning…
DM: There are some things you have to practise. And there's tacit knowledge that you can't pick up from just talking and reading. Agile Learning won't help you learn to ride a bike. But there's the theory test that's now part of the Driving Test: you pick these things up not just through the British School of Motoring, but through talking to your friends, going on a simulator, buying a book and so on. Most of the Day Skipper stuff is made out of practical elements that you get ticked off and signed when you've demonstrated you can do them. For the theoretical stuff there is also a summative assessment, and our instructor used his discretion to identify the things where he said, "You've just got to know this."These are examples of Agile Learning blending into accreditation.
DJ: What is about this learning that makes it agile?
DM: You define the pace, you decide which elements you want to do when, you decide who you learn with as a group. You can do this Day Skipper certification in a week's crash course — and you pay a fortune! What we did a weekend taster, and we got some elements of our "competent crew" accreditation there. Then we looked at the syllabus, and did the theory learning over the winter, through a blend of classroom learning and working through online instructional materials, which made it possible to make up for the face-to-face sessions that I missed.
It was agile to me because I had a suite of tools, a suite of communication practices. Our instructor had one of the highest success rates in the country, and I think the way he achieved this was to give people different methods and tools, and he attacks risks first. He said he was going to the assessment of the first theory part after four weeks, but actually did it after six. There was a real blend of formal and informal learning: we went to the pub afterwards and were able to explore issues in a different way then.
Agile Learning doesn't have to be assessed, but if you want it to be, you can put in methods that will work.

Where do you think Agile Learning might fit with the roll out of mobile internet offerings?

Mobile internet, especially with things like Android phones is to my mind the most exciting new platform since the Sinclair spectrum and I expect it to have a greater impact than the PC or laptops. Combining a whole range of sensory devices (light sensor — camera; directional — compass; sonic — microphone; RFID), combined with GPS and internet connection these devices have a capability that will open up significant new learning opportunities. We can in effect carry a learning device that will increasingly understand its physical context and allow us to integrate our world. With applications like ZXing andGoogle Goggles we already have the ability to analyse photographs, extracting semantic information and other data from them, and linking through to secondary sources.Increasingly the world through which we navigate will contain a data layer from which meaning can be accessed and knowledge inferred.
Increasingly we can expect information and knowledge to remain in the cloud rather on personal or corporate servers (my feeling is that mobile internet will accelerate this) this will provide even larger layers of data that will be mined for meaning.

What areas do you think mobile and Agile Learning solutions might be most useful — either in the developed or developing world.

My feeling is that the entry cost and thirst for knowledge and agile how-to knowledge will drive the emerging economies to adopt Agile Learning solutions quickly. Access to first generation mobile signal technologies is rapidly becoming ubiquitous when measured against population density providing an communication infrastructure that is changing economic models not altered for centuries.
There is a thirst and demand for education, and knowledge at a low unit cost outside what we call the first world that these technologies may be able to provide fulfil. Closer to home, we can may see mobile internet blur the boundary between formal and informal learning, access to knowledge and information becoming always on.





Agile Learning and Agile Software Development; A Discussion
Dick Moore was Director of Technology at Ufi/learndirect and has a deep understanding of technology infrastructure and agile development methodologies and reflects on how this may affect agile learning practice. These are laid out in the table (DAVID REF)
We can build agile learning experiences, in an evolutionary fashion. You set out with a clear goal, and iterate as you go to make sure that you're still on track. In the process you learn something; perhaps that this is the wrong way to do it, or that you're asking the wrong question! Having clear goals, which are easily tested so you know when you've reached them, does not mean that you can't change the goals, with the consent of the interested parties.

Shallow v Deep Learning & Core Agile Skills
The need to rapidly acquire new skills and knowledge, combined with the knowledge engine that is the internet, promotes self-directed learning — be it formal, informal or recreational; Access to how-to videos via YouTube, and the previous set of how-to documentation that underpinned the open source Linux development platform have shown that recipes, plus a learning goal, can form the basis of significant learning and development programmes, especially if undertaken through a series of learning iterations.
There is a view that says that every Google query is a piece of shallow Agile Learning; there is a goal (a question requiring an answer), a tool (search engine) and a need for some analysis of content (the results returned by the engine). However you need to be capable of discriminating on the reliability of sources, so effective use of search engines may require a range of core agile skills, which could be;
a) formulation of queries,
b) how to judge the providence and veracity of what is returned
c) basic technology skills that are not yet universal.
The degree to which these core agile skills are similar to the literacies of self-organised learners that David Jennings blogged about is more about your degree of competence then whether you have these skills or not. Anyone who uses Google will have them to some extent. Having a clear goal that one or more people can focus around in short iterations and a way of measuring this provides an end point; clear criteria that define the end of a learning iteration can only be a good thing.
These might be;
• after this iteration I will be able to…
• after this iteration I can demonstrate…
• after this iteration I can explain and show the relationships between…
So having these core skills and an understanding of learning iterations are key for agile learning.

Assessing Agile Learning
The assessment challenges with Agile Learning are much the same as with any other form of learning. If certification is required, then there has to be some sort of strong and rigorous assessment that underpins the knowledge and practical skills being taught.
Concerns such as impersonation, plagiarism, weak testing regimes, and corruption all apply. However in Agile Learning we have some advantages in that clear goals are set, intervals are defined and are typically short and there may be an end result delivered via a group allowing the group to self-assess (something that should be encouraged). Moderation may of course be needed.
For more formal qualifications then formal assessment might be considered. The UK Driving Test with its theory and practical test might be considered a good example of Agile Learning. There's the theory test that's now part of the Driving Test: you pick these things up not just through the British School of Motoring, but through talking to your friends, going on a simulator, buying a book and so on. This is an example of Agile Learning blending into accreditation. There is also ‘tacit knowledge’ that you can't pick up from just talking and reading. Agile Learning won't help you learn to ride a bike.
In my experience in Agile Learning you define the pace, you decide which elements you want to do when, you decide who you learn with as a group. It is agile to me when you have a suite of tools and a suite of communication practices allowing a real blend of formal and informal learning. Agile Learning doesn't have to be assessed, but if you want it to be, you can put in methods that will work.
Agile Learning and the roll out of the mobile internet
Mobile internet, especially with things like Android phones, is the most exciting new platform since the Sinclair spectrum and I expect it to have a greater impact than the PC or laptops. Adding a whole range of sensory devices (light sensor — camera; directional — compass; sonic — microphone; RFID), combined with GPS and internet connection, will open up significant new learning opportunities. We can in effect carry a learning device that will increasingly understand its physical context and allow us to integrate our world. With applications like Google Goggles we already have the ability to analyse photographs, extracting semantic information and other data from them, linking through to secondary sources. Increasingly the world through which we navigate will contain a data layer from which meaning can be accessed and knowledge inferred.
We can readily expect information and knowledge to remain in the cloud rather than on personal or corporate servers, my feeling is that the mobile internet will accelerate this, providing even larger layers of data that will be mined for meaning. Outside what we call the first world there is a demand for education, and a thirst for knowledge at a low unit cost, that mobile technologies may be able to provide, so the emerging economies might adopt Agile Learning solutions quickly due to the low entry cost. Closer to home, we may see mobile internet blur the boundary between formal and informal learning, access to knowledge and offering information that is always on, encouraging agile learning practices.