RSS

Category Archives: Robotics

EduTech Africa 2019 – Coda

Last week I attended the EduTech Africa 2019 Conference in Johannesburg and would like to wrap up my thoughts on the conference with a few observations. Now that the dust has settled the thing that sticks out most in my mind is the clear recognition of the rise of Computer Science as a K-12 academic discipline. The government’s commitment to rolling out IT as a subject, and the focus on coding across all age groups has established a clear sense that Computational Thinking and Computer Science belongs in the core curriculum in all schools. The big question is then how we get there. The announcement recently that PISA Assessments, which offers international benchmarks in Maths & Science, will now include Computational Thinking and Computer Science is confirmation of this. Most of the talks I attended addressed the issue of how best to teach Computer Science in some form or other. Passionate teachers shared their best practice, and their failures. So, the coda to my reflections on the conference is really to address that question. Is there a best method to teach Computer Science?

NS Prabhu (1990) in answering the question of whether there is a best method of teaching or not, concluded that the key factor in teaching success lay with the teacher’s sense of plausibility, the teacher’s sense of self belief that what they are doing makes sense, how passionate they are. There is clearly a great deal of plausibility around the teaching of Computer Science at the moment. Obstacles are being dealt with as opportunities, and there is a very real sense that inventiveness and creativity can overcome the constraints of budget and lack of training.

The clear consensus amongst teachers seems to be that physical computing forms the best approach. Most presentations highlighted the use of coding in conjunction with 3D printing and robotics. My very first exposure to teaching computing was with Seymour Papert’s (1980) logo system. I did not have the turtles, using only the computer interface, but I tried to make it more concrete by using physical cards with shapes students had to emulate. Computer Science is a very abstract subject and needs to be concretised for students as much as possible. The cost of all the kit needed to do this is prohibitive.

I recently came across micro:bits which uses a web-based platform for coding. The code created is then downloaded as compiled hex code to the microbit chip which executes the code. But crucially it also has a web visualisation tool, which executes the code in the code editing window. The micro:bit controllers are themselves fairly cheap, but having a visualization tool means that more students can code at any one time. A class would need fewer physical chips at any one time. I have not yet been able to test the real thing, but it seems to me a perfect fit for the kinds of physical computing tasks I would wish to introduce. It uses a block coding interface, but you can toggle to program in JavaScript or Python, making it ideal for transitioning between block-based coding to the text-based fare students will need higher up the school. You can also design 3D printed parts for interesting projects.

But I digress, back to best methods. Another strong thread in the conference was computing for problem solving. I have to say that I am a little dubious about the whole Computational Thinking leads to better problem solving generally. I believe it leads to better problem solving in computational contexts, but transfer of skills from one context to another is always problematic in my view. Nevertheless, I do believe that students should be given real world problems to solve as far as possible and Computer Science teachers are leading the way in envisioning how coding could form a central plank in cross-disciplinary problem solving exercises. There was a great deal of talk at the conference about the need for teachers to “come out of their silos.” There is certainly no need for CS teachers to set projects divorced from the real world, or set problems narrowly about computers.

The final method that was raised at the conference was unplugged computing, an approach which involves modelling algorithmic thinking without a computer. For example students might be asked to write code to control a class-mate acting as a robot to perform a certain task. A talk by a primary school teacher on coding in the junior years had us all playing rock, paper, scissors. I’ve forgotten why, but it was great fun!

 

In the end, my take-away from the conference was to think about the best approaches for my own classes. And most particularly how to integrate all three of these approaches better. To my mind this is the best sort of take-away!

 

Bibliography

Papert, S, 1980. Mindstorms : Children, Computers, and Powerful Ideas. Basic Books. https://dl.acm.org/citation.cfm?id=1095592.

Prabhu, N.S, There Is No Best method – Why?, TESOL Quarterly, vol. 24, issue 2 (1990) pp. 161-176
 

Coding & Robotics Summit – Johannesburg

Artificial Intelligence and the Fourth Industrial Revolution have become buzzwords in education, often used with little thought or understanding in slightly Pavlovian ways. There is a very real sense that big changes are afoot, and everyone is nervous about how to respond, and most particularly to be seen to be responding. I am not quite sure what to make of it, frankly. While it is clear that the working landscape will change as a result of AI, I am not convinced that much in the educational field changes … until it does. What do I mean? I think that it is already crystal clear that education should be looking to teach critical thinking, collaboration and creativity. I’m not sure that anything has changed around this. It is my belief that we should have as broad a curriculum as possible. Drama, History, Music should be as core to our curriculum as STEM. It makes no sense to me to de-emphasise or over-emphasise any field. So while I am all in favour of ensuring that coding & robotics forms part of the curriculum, I find the whole STEM, STEAM, and now STREAM (with robotics) debate counter-productive.

There will come a point, however, at which Machine Learning is powerful enough that meaningful AI applications are ready for classroom implementation. When Watson and Skinner built their teaching machines in the last century they imagined programmed learning which allowed for instant feedback and personalised learning paths, the kind of thing advocated by Pestalozzi back in the late 1700s with his one-on-one tutoring methodology. What emerged though, was the kind of drill and kill learning platforms that are the kiss of death for education. Computers are simply not intelligent enough to be able to spot when students are gaming them. However, AI does offer a possible resurrection of the idea with systems that are far more responsive and capable of analysing student production with sufficient nuance as to be useful. Real-time feedback loops enabled by devices which can read the learning brain, are not Science Fiction anymore. Nor are teaching machines which can sift the huge amount of data collected and make sense of it. There will undoubtedly, then, come a point at which AI teaching machines enter the classroom. In the lead up to that we can probably expect a range of apps that employ AI in some way, and answer particular pedagogical needs. As I get older, face and name recognition would be nice! But full-blown AI in the classroom is a little way off yet. Meaningful data analytics is probably much closer, but I’m not convinced having a wealth of data is always a good thing. I am also afraid that that data will be harvested for purposes unrelated to education. Imagine how you could Cambridge Analytica a population if you owned the data being collected on how everyone learns and thinks?

The focus of this one day conference was on Computational Thinking and on coding and robotics as a vehicle for teaching thinking skills, building the habits of mind and dispositions necessary for a post-singularity world. The South African government has recently announced that it will be introducing coding into the primary and GET phase (middle school) curricula. Karen Walstra opened affairs by walking through the history of Computational Thinking and its component parts, how coding concepts can be used across the curriculum and not just in coding classes. I am planning a blog article on Computational Thinking so I won’t dwell on it here. Her talk was vital in terms of introducing Computational Thinking, and in laying thinking skills as the foundation of any curriculum changes. However, what worries me is that an increased interest in coding has elevated it beyond what I think it is capable of providing. Computational Thinking, Coding & Robotics is not a magic bullet which will suddenly solve all education’s ills. It is a necessary skill to learn, a useful knowledge base, and a set of dispositions that all students need, but it should be seen as forming part of the thinking skills programme, not replacing the existing curriculum. All subjects and all skills are vital, in different ways. Don’t get me wrong, I am all in favour of coding’s place in the curriculum, but learning problem solving skills requires a broad world knowledge, and there are a number of thinking skills beyond the computational that are needed. Important, yes. A magic bullet, no.

St Enda’s Secondary School students designing a school website, circa 2003.

What her talk did highlight was the notion that all students can benefit from learning coding. The team from CodeJIKa presented a cogent case for this with their wonderful extra-curricular code club programme teaching students HTML, CSS and a little JavaScript. They have an online curriculum which runs largely through peer to peer learning. When I was teaching Computer Applications at St Enda’s Secondary School in the early 2000s I used the same approach. HTML & JavaScript are browser based and so do not need compilers and can be used offline – a huge consideration where Internet connection is a big problem. To my mind starting with a markup language is also helpful because it is easier to slip into, helping students get into the habit of moving between the concrete and the abstract. You can then start to slip JavaScript in quite organically and start introducing key programming concepts. Robyn Clark, from CodeJIKA, stressed how web design is also helpful in building entrepreneurial skills, giving students a side hustle. The CodeJIKA approach is to my mind a fairly easily replicable model across under-resourced schools. It is also flexible and stackable as App development, robotics and programming proper can be added as the skills and knowledge base increases. Amini Murinda from ORT South Africa presented what they have been doing in expanding coding and robotics in a growing number of schools. Both programmes clearly show that coding & robotics initiatives are engaging and transformative.

We heard from two speakers representing robotics companies, who spoke about where robotics and AI is headed, and why we should not fear job losses, and how investing in coding in primary schools could reduce failure rates in higher education. These talks provided a useful backdrop and a perspective from the world of work. I would have liked a greater emphasis on curriculum, but the summit was useful in bringing together participants from industry, teacher training and secondary and tertiary teaching sectors. It would have been great if Government had also been represented. We need many more of these discussions.

The big take-away for me was the need to take these pilot projects, together with the experience of primary and secondary teachers from the private sector who have been developing their own programmes, share best practice and work on a curriculum and pedagogies that make sense.

 

EduTech Africa 2018 – Moving Beyond the Technology to Make a Difference t

Over the last decade or so the focus of the ed tech conferences I have attended has shifted increasingly away from the technology itself towards what we can do to transform education. In the early years it was as if ed tech enthusiasts were like magpies, dazzled by every shiny new tool. Some of that sense of wonder still exists, of course, and is healthy. We need to be alive to new possibilities as technology evolves. But over the years we have learned to become more discriminating as we found what tools actually worked in our classrooms, and learned not to try to do too much at one time. The focus started shifting towards pedagogy, towards how to use the tools effectively. Behind this was always some thought as to the significance of the impact of technology on education. Common refrains have been the development of 21st Century Skills, personalised learning, a movement away from teacher-centred to student-centred approaches, problem-based learning, what technologies will disrupt education and learning based on the burgeoning field of neuroscience. The overall sense has been one of promise, that technology has the potential to make teaching and learning more effective, and that education will become transformative in liberating humanity from a model  grounded in the factory system and a mechanised reproduction of knowledge and skills.

 

This year’s conference was no different in content although the technologies have changed somewhat. The focus has shifted towards Artificial Intelligence, robotics and coding, especially how to involve women in STEM and how to infuse computational thinking across the curriculum. However, this is the first time the sense I have is not one of advocacy, but of militancy. Speakers from the world of work were united and adamant in a condemnation of schooling itself. A clear preference for extra-curricular learning and the futility of academic qualifications was presented stridently. Employers, we were told, prefer people able to solve problems. If any learning is required it can be delivered, just-in-time at the point of need, online via MOOCs. Tertiary qualifications should be modular and stackable, acquired over time when required to solve real world problems. Educators endorsed this stance stressing personalised learning and the use of Artificial Intelligence and even real-time feedback from brain activity. The sense was one of an urgent need for a curriculum based on problem solving rather than subject disciplines. If you need some Maths to solve a problem you can get it online. You don’t need to study Maths divorced from real world imperatives.

 

The very idea of tertiary institutions is clearly under massive assault, and it cannot be long before they come for secondary schools as well. What scares me about this is not that I don’t agree that learning should be problem-based at some level, or that degree programmes should not be using MOOCs and blended models to achieve greater modularity and be more student-driven. What scares me is what we lose by doing that. My fears are based on two premises.

 

Firstly, I believe that knowledge should be pursued for knowledge sake rather than for the needs of the world of work alone. Of course our education should prepare us for employment or entrepreneurship. To argue that it shouldn’t is folly. But knowledge has its own trajectory and logic. Mathematical knowledge, for example, represents a body of knowledge bounded by rules and procedures. It forms a coherent system which cannot be broken up into bite-sized chunks. Can one quickly study calculus without studying basic algebra just because you need calculus to solve a problem? Historical knowledge is not just about reading up on Ancient Sumeria on Wikipedia quickly. Historical knowledge is founded on a system of evidentiary inquiry within a narrative mode of explanation. I worry that just-in-time knowledge will lack a solid enough base. If we erode the autonomy of the universities and do away with academic research, what happens to knowledge? It will become shallow and facile.

 

Secondly, I believe that the discovery model of learning is deeply flawed. Of course, if left to our own devices, following our curiosity, we can discover much. It is a fundamental learning principle. But it is not very efficient. There is no earthly reason why teaching should be ditched. Being told something by someone else is as fundamental a learning principle as learning something for yourself. It is an effect of socialised learning. We learn from each other. Teaching is an ancient and noble profession, and there seems no reason to ditch it now. The scholar’s dilemma is that it is unusual to discover anything unless you know it is there, and this requires guides and mentors. The world we live in is complex and vast and we need a working knowledge of a great deal. Without extensive teaching, it is difficult to see how we could acquire the knowledge we need.

 

I would argue that we need a broad-based liberal education, focusing on critical thinking and problem solving, which gives us a grounding in Mathematics, the Sciences, the Arts and Humanities. At this stage, after a first degree, say, the best approach could well be just-in-time content delivery delivered online.

 

Just because technology can disrupt education doesn’t mean it should. Teachers have been very conservative in their adoption of new technologies, and I think this is a good thing. Education and knowledge are just too important to change willy nilly. We need to be certain that we are not destroying our evolutionary advantage, our ability to think, simply because we can.

 

EduTech Africa 2018 – Day 2 of Just-in-time Learning

 

Dr Neelam Parmar

On the second day of the Conference the focus seemed to shift from what schools should be doing, to the nature of learning itself. Dr Maria Calderon took us on a whistlestop tour of what neuroscience has to tell us about learning. Key to understanding this is the surprising role played by emotion in mediating learning experiences. If the amygdala is too excited learning is blocked. Ian Russell then stressed the importance of changing the way learning happens in schools so that it reflects how the world now works and students are better prepared for the world of work. Learning needs to be flexible and delivered just in time. Employers are interested in your skills not your qualifications. The days of students earning a degree and then entering the world of work are gone. Mark Lester amplified this idea by stressing how tertiary learning is increasingly blended and modular. Life-long learning is the new norm.

Dr Neelam Parmar presented us with a model for weaving together technology and pedagogy. Choices around technology and pedagogy are driven by decisions around curriculum and finding a match between schools and the world of work. She left us with an image of the accelerated use of AI in schools: robots in China that monitor student attention and nudge them awake when they fall asleep.

It is in many ways an image which encapsulates the future and its possibilities. Technology can deliver a more personalised, seamless tracking of educational achievement, much of it delivered online. Students of all ages can learn what they need to learn just in time, building their own curriculum. The curriculum can be based on the task, the challenge at hand. And yet there is a danger, a danger that we will lose the ability to discriminate out what it is that is important to learn. The dilemma of self directed study is that you can’t know what you need to learn until you have learned it.

There is a strong movement away from traditional school disciplines, towards problem based learning, and I believe this is a mistake. Knowledge is coherent because it is bounded by a field. If it becomes nothing more than fodder for solving problems we lose something very valuable and that is the pursuit of knowledge for knowledge sake. Something happens when you do history for its own sake, not just to prepare for a career in politics, for example. Or if you do maths just for engineering. You lose a certain perspective, you lose knowledge itself. Knowledge is not just something you gain to live, it is something, almost tangible that enriches our lives because it throws up surprising perspectives and unleashes powerful forces of change.

The conference this year had a strong sense that the teacher is increasingly irrelevant, and I’m not that convinced that wide awake robots are the best solution. I think the teacher will be with us for quite a while yet!

 

 

EduTech Africa 2018 – Day 1 Just-in-time Teaching


The first day of the Conference started with an impassioned plea from Sameer Rawjee to make schools places where possible futures could be prototyped rather than relying on the reproduction of the present. He envisioned a future world of technology where the role of technology was to make our lives easier and liberate humanity. Schools should be places where this vision of a future where humanity has a place and can thrive is fostered and explored. This set the tone for a conference where coding, robotics and Artificial Intelligence was foregrounded, and where the role of technology was to transform pedagogical practice, empowering flexible, life long learning focusing on the development of skills, attitudes and dispositions in tune with a changing world.

Chris Rodgers spoke next on robotics and the importance of makerspaces in fostering learning and problem solving as a basis for integrating and reorganizing the curriculum. When solving problems, students arrive at a diversity of solutions, and draw on what they need to know, when they need to know it. Teaching becomes just-in time interventions, reflecting the way the world works.

In the break away sessions this theme was amplified. The role of the teacher has to change. Learning needs to become more flexible, and with this change comes the need for relevant knowledge on demand. A move from a push to a pull model, if you like. The classroom of 2030 will have to reflect this out we will have failed or students.

 

Using Algorithmic Thinking to Teach Writing

The gains being made by Artificial Intelligence are truly impressive, but we may not be at the stage where a robot can out-write Shakespeare. And yet I do believe that we can use algorithmic thinking to teach students to become better writers. One of the bug-bears for many students over the years has undoubtedly been the lack of explicit instruction in how to write. The dominant pedagogy has been to give students plenty of opportunity to practice creative writing, and to attempt to mold improvement through feedback – often woefully inadequate feedback.

This image was originally posted to Flickr by Scoboco at https://www.flickr.com/photos/62159569@N08/10546981384. It was reviewed on  by FlickreviewR and was confirmed to be licensed under the terms of the cc-by-sa-2.0.

And yet writing can clearly be taught. At the very least students should be made aware of the overall structure of any piece of writing: how to set out a clear thesis statement and develop ideas in successive paragraphs which develop topic sentences, fleshing each idea out with anecdote, fact or quotation. If they are practising these skills quite explicitly their practice is focused and directed, it is far more likely to bear fruit.

I have found, over the years, however, that no amount of scaffolding will make this process easy to implement in whole class instruction. Most students can use conceptual maps to plan a sequence of ideas which support a thesis, but really struggle when it comes to developing these ideas in individual paragraphs. For some this process appears to come naturally. They effortlessly weave together anecdotes and observations to illuminate their ideas. Others appear incapable of marrying abstract ideas to concrete details, which is really what is at stake here.

I had a brain-wave the other day in a coding class. Could the kind of thinking used in coding (algorithmic thinking) not be employed to help bridge the seeming chasm between abstract and concrete? Algorithms, after all are sets of instructions which a machine can follow – a blue-print for successful practice. Maybe, for those who seemed to lack a muse, a blue-print was exactly what was required. And maybe, after following an algorithm for a while, the patterns and habits might stick.

I started by getting the whole class to generate a set of ideas and sub-points using a graphic organiser. We then organised these ideas into a logical sequence so that an argument was constructed. I put these ideas on a Google doc which I then shared with the class on Google Classroom so that each student had their own copy. The class then broke into groups and had to find anecdotes, facts, details or quotations. These were written out on cards and shuffled into a pile. Students were then told to start writing, using the logically sequenced outline we had developed. As they wrote each paragraph they had to come to the front of the class, dig in the pile and try to find at least one anecdote, fact or quote to use in that paragraph. When they had finished they returned it to the pile.

Individual essays were thus unique. The same anecdote could be used to support or refute an idea. We then shared some of these sequences in essays and discussed how they had been used to support the thesis statement. My follow-up, next term, will be to get students to select ideas from a pile and match these to the developing details so that each essay has a different sequencing structure.

 

EduTech Africa 2017 Day 1 – The Search for Soft Technologies

DLMgdPwW0AEaJSG

I am once more attending the EduTech Africa Conference and I like to try and distill from the presentations and conversations at the Conference a sense of where we are sitting with Educational Technology in South Africa. As usual there is a narrative, repeated almost like a mantra, around the desire that technology will transform educational practice and deliver a more student-centred curriculum and pedagogy. Margaret Powers delivered a powerful keynote which summed this up succinctly and persuasively. There was an air of optimism this year which replaced the more messianic tone of previous years. Maybe it’s a sense that that goal is a little nearer, a little more achievable. Or perhaps it’s just that there is generally a new optimism abroad, despite the election of Trump, a return to the politics of hope reflected in the rise of Corbyn and Sanders, a sense that no matter how massive the task, the monolith of schooling can be re-imagined and re-envisioned just as there is a sense that the bastions of the political establishment can be assaulted.

But the highlight of the day for me was Stephen Heppell’s address. Heppell’s work on re-imagining the architecture of the school through the design of learning spaces that offer affordances for the kinds of educational transformation the Conference is calling for, is legendary, but it found a particular resonance this year with the track that I followed, that to do with the rise of coding and robotics. This link between designing classroom spaces and coding may seem tenuous, but ultimately it is about the locus of agency. Heppell spoke about the need to give agency to students through the design of learning spaces, and coding and robotics gives the same agency over machines.

Technology may be considered hard or soft, and what I mean by this is the ability to be flexible. Hard technologies do not alter easily, they are solid and fixed. School buildings are hard technologies. You cannot just knock down a wall to accommodate extra students. Soft technologies are flexible and versatile, they can be re-imagined and re-purposed on the spot. Teacher’s pedagogies are soft technologies, in that they can be changed at a moment’s notice to suit what is happening in the classroom. This is why teachers normally change their pedagogies to suit the space they are in. Teach in a lecture theatre and anything but teacher-centred pedagogies are nigh on impossible! As a teacher I have twice been the incumbent computer teacher when the computer room was re-designed, and not once was I or my students consulted! I loved Heppell’s insistence on building learning spaces to children’s specifications!

To my mind what coding and robotics offers is a soft curriculum to replace a hard curriculum, a curriculum based on problems defined by the students themselves, to which solutions are sought collaboratively. Marina Myburgh’s presentation on her exploration of a coding and robotics syllabus at Crawford, Sandton defined for me the journey many schools will be taking over the next few years as we seek to replace our now out-of-date computer skills syllabi with a new curriculum which seeks to map out how computational and algorithmic thinking can enhance all learning. There was a remarkable sense of purpose in the coding and robotics round table discussion that allowing students to explore solutions to problems they define is the way forward. The task is now to research and explore the optimum learning paths to achieve this.

This will involve not only teachers of Computer Skills, for coding & robotics extends across the curriculum. The focus on STEAM and the Maker Movement emphasizes the extent to which we need to ensure that a future in which AI and robotics increasingly threatens our job security needs to be tempered by a concerted effort to ensure that we as human beings are able to retain some control, some agency in our lives. Coding and tinkering may be the most liberating and humanist of all the academic disciplines as the 21st Century starts to get a grip!

 

 

 
 
%d bloggers like this: