THE MAGAZINE FOR COMPUTING

& DIGITAL MAKING EDUCATORS

IN DEFENCE OF PSEUDOCODE
Evidence for its use in education

LEARNING GRAPHS
Maps that show progression

INCLUSIVE BY DESIGN
A shift in thinking to meet all your learners’ needs

Issue 12 Apr 2020 helloworld.cc

MACHINE
WHY STUDENTS SHOULD MASTER
THE INNER WORKINGS OF AI

LEARNING

FROM TURING BUTTERFLY NATURAL LANGUAGE

TO ALPHAGO OR BIRD? PROCESSING
Inspire your students Classification algorithms Block-based coding activities
with the history of AI demystified to explore the tech

PLUS PICTURE
•
INSIDE AN OFSTED DEEP DIVE EVERYDAY ASSESSMENT APPROACHES PROJECT-BASED LEARNING
1 (helloworld.cc)
•
BOOKS IN COMPUTING EDUCATION • HIGHLIGHTS FROM WiPSCE • COMPUTING POETRY
The Gender Balance
in Computing programme
Are you a Computing teacher based in England?
There’s an important new programme that your school can
participate in…
The Department for Education recently funded our Gender Balance in Computing
(GBIC) programme, which gives us the amazing opportunity to investigate
different approaches to engaging girls in computing in primary and secondary
schools. Our research will also be used to find out how we can increase the
number of female pupils who select Computer Science at GCSE and A level.

We want to create the widest possible community of schools to help us try

different approaches and learn from what we find out. That’s why we are
inviting all the primary and secondary schools in England to take part.

We’d love your school to participate. Join our Schools Network today and help us
to find out what really makes a difference to getting more girls
into computing.

Find out more: rpf.io/gbic-info

 EDITORIAL
SUBSCRIBE
IN PRINT Editor
FOR FREE Sian Williams Page
TURN TO PAGE [email protected]
34
HELLO, WORLD!
Sub Editors
Joanna Garbutt and Louise Richmond
Subscriptions
Joshua Crossman
achine learning and AI have an teaching the inner workings of this technology. It’s

M almost unfathomable reach into our

lives. From everyday applications
great to share some of their brilliant work in this
issue of Hello World. DESIGN
such as scrolling through social media or using Last issue, we had a cover theme of inclusion criticalmedia.co.uk
GPS, to the large-scale collection of citizens’ and diversity. Reflecting on the emails we Head of Design
data for corporate or political aims, this received and feedback in our survey, I’m really Lee Allen
technology affects us all. And if the goal of happy to say that Catherine Elliott will be Designers
teaching computing is to put the power of continuing the discussion around this in her Harriet Knight, Sam Ribbits, Ty Logan
technology into young people’s hands, where new column. Photography:
they can understand and control it, rather than On page 95 we’ve included some of the Jonathan Howard, Brian O’Halloran and
be passive users of it, then surely we must feedback from our 2019 reader survey, and how Fiacre Muller, Raspberry Pi Foundation
teach machine learning. we are responding to it. Adobe Stock Photo

This is challenging, though, due to the sheer If you have feedback for Cover
complexity of many of machine learning’s us, or an idea for a Rob Jervis

underlying algorithms. Even David feature, opinion piece, or

Spiegelhalter, a mathematician whose work in activity that we could CONTRIBUTORS
the 80s underpins a lot of machine learning include in the magazine,
Adrienne Tough, Adrienne Webb, Alan O’Donohoe,
today, admitted to me he would struggle please do get in touch. Alexandros Zeakis, Ben Hall, Brent Richardson, Bruce
following by hand every step of the calculations Robinson, Carrie Anne Philbin, Catherine Elliott, Dale
Lane, David Spiegelhalter, Eirini Kolaiti, Emily Driemann,
in many of them. Despite this, many teachers Sian Williams Page Estelle Ashman, George Boukeas, Hayley Leonard, Helen
and organisations are not shying away from Editor O’Kelly, Janina Ander, Jesualdo Martinez, John Woollard,
Katie Molloy, Ken Kahn, Kevin Johnson, Manos Zeakis,
Marc Scott, Mark Calleja, Michael Conterio, Michael Jones,
Nazia Fakhruddin, Neil Rickus, Nicholas Hughes, Oliver
Quinlan, Pat Seaver, Richard Millwood, Sian Williams
Page, Stew Edmondson, Tayler Kirk, Thom Kunkeler, Tim

FEATURED THIS ISSUE Bateup, Ursula Martin

Hello World is a joint collaboration:

This magazine is printed on paper sourced from

sustainable forests and the printer operates an
environmental management system which has
been assessed as conforming to ISO 14001.
Hello World is published by the Raspberry Pi
KEN KAHN NAZIA FAKHRUDDIN URSULA MARTIN Foundation, 37 Hills Road, Cambridge, CB2
1NT. The publisher, editor, and contributors
accept no responsibility in respect of any
Ken is a researcher in the Department Nazia is a Royal Institution Computer Ursula Martin has been an educator omissions or errors relating to skills, products
or services referred to in the magazine.
of Education at the University of Oxford. Science Masterclass Speaker and for 17 years. She is a District Level Except where otherwise noted, content in this
He was a member of Seymour Papert’s Code Club Volunteer for at a library in Technology Resource Teacher in magazine is licensed under a Creative Commons
Attribution-NonCommercial-ShareAlike 3.0
Logo Group at the MIT AI Lab. the London Borough of Hounslow. Mobile, Alabama. Unported (CC BY-NC-SA 3.0).

helloworld.cc 3
 CONTENTS
COVER FEATURE
15 19

POPBOTS
An innovative project run by MIT

20

THE HISTORY OF AI
Fascinating stories to share NATURAL LANGUAGE PROCESSING
with your students Explore the tech behind Alexa with Snap!

NEWS, FEATURES, AND OPINION

6 NEWS 22 WiPSCE 54 CESI•CS

EdTech’s response to COVID-19, Highlights from the conference News from the grassroots
Coolest Projects USA, and more voluntary organisation

36 OFSTED TIPS
12 INTERVIEW: DAVID Inside a computing deep dive
56 PROJECTS IN CLASS
SPIEGELHALTER Students take charge of their

42 CRACKING IDEAS
The statistician on data science in own learning
schools and the limits of AI An invention competition for
58 DEEPER LEARNING
ages 5 to 11
14 THE INCLUSIVE CLASSROOM A pedagogical approach for
43 DIGITAL SCHOOLHOUSE
A new column from Catherine Elliott meaningful experiences
Play-based learning workshops

30 STORYTELLING 62 LEARNING GRAPHS

Teaching computing concepts with
picture books
48 CODING & 21ST-CENTURY SKILLS The English computing curriculum
mapped for progress
The importance of a safe space and a
sense of belonging in the classroom

32 DEVELOPMENTS IN TECH 66 ONLINE TRAINING

Ensuring there’s opportunity for all in
50 PSEUDOCODE Behind the scenes of Raspberry
computer science How it can support learning Pi’s courses

4 helloworld.cc
 CONVERSATION
92 YOUR QUESTIONS
The Raspberry Pi team answer your
queries and offer their thoughts

LETTERS
26
MACHINE LEARNING 94
AT PRIMARY SCHOOL This issue: is gaming in the
How to introduce the classroom a good thing?
topic to young learners

95 SURVEY
Feedback from our readers at the
end of 2019
AI IN THE CLASSROOM
Resources to structure discussions

23
28
REVIEWS
97 BOOK REVIEWS
An updated guide to body image

24

APPS FOR GOOD

Courses that encourage under-represented
YOU AND AI
Engaging activities
LEARNING
groups to study computing from the Royal Society
RESOURCES & LESSON PLANS

68 PRACTISE WITH POETRY

Help with computing vocab

71 HOLOGRAPHIC ILLUSIONS
39
SCRATCH FOR LANGUAGES
Block-based programming 3D programming
in MFL lessons
74 INTRODUCING PYTHON
An interactive lesson and workbook

76 MAX’S QUEST
A pirate machine learning adventure
from a CoderDojo

80 MACHINE LEARNING FOR KIDS

Resources for primary-aged students

83 BUTTERFLY OR BIRD?
Introducing neural networks

ELECTRONICS EVERYWHERE 86 AI AND ETHICS

45 A new university outreach
scheme
Thought experiments and the moral
dilemmas of AI

helloworld.cc 5
NEWS FEATURE

CORONAVIRUS OUTBREAK SEES

A SHIFT TO ONLINE EDUCATION
The widespread implementation of online learning is taking place at an
unprecedented rate, with impacts on student attainment and teacher well-being
Sian Williams Page

s the number of Covid-19 Which platform, and how to use it? children’s virtual learning. It’s a tough
A cases continues to rise globally, In many schools, the task of choosing and balance that, on the first day of virtual
around 372 million children are currently managing the correct online platform to learning, has already fried our nerves.
experiencing an upheaval in their education use in the event of a closure falls on the Between Seesaw, the emails I’ve received
due to school closures. “The scale and shoulders of the Computing teachers. from our kids’ various and wonderful
speed of the current educational disruption In the absence of more detailed current teachers about Seesaw links, and my own
is unparalleled and, if prolonged, could guidelines, teachers are facing challenging Flipgrids, email, Google classrooms, Google
threaten the right to education,” said decisions around which technology to chats, and PowerSchool learning sites, I am
UNESCO Director-General Audrey Azoulay. use and how to go about implementing already overwhelmed.”
Schools are turning to distance learning it. Many ed-tech companies are now There are concerns that moving to online
programmes in an effort to continue offering their products for free. UNESCO learning could increase the attainment gap
providing an education for their students. has published a list of nearly 30 free between richer and poorer students. In the
In the UK, most schools are open as programmes that they say have “wide UK, seven per cent of homes do not have
of 17 March, but teachers are taking reach, a strong user base and evidence access to any kind of internet connection,
measures to prepare for widespread of impact”. and many students do not have access to a
closures. James Bowen, director of policy School closures have brought extra laptop or tablet. Many children do not have
at the NAHT school leaders’ union, told Tes challenges to teachers who are parents a suitable place to work at home.
that his organisation is working with the themselves. Stan Covington, a teacher in There are also serious concerns about
DfE to “ensure they consider all scenarios Bangkok, has described the challenges of students going hungry due to school
and possible eventualities... We are urging implementing various online systems closures. In the US, 20 million meals are
them to continually review and update their while juggling childcare. “My wife and distributed for free in schools each day.
advice to schools. This includes scenario I, both teachers, are struggling with The most recent data available suggests
planning in the event of regional or national balancing the demands of virtual schooling around 14 per cent of pupils in England
school closures.” and the demands of managing our young claim free school meals.

6 helloworld.cc
 NEWS FEATURE

LANGUAGE SKILL A BETTER

PREDICTOR OF CODING
ABILITY THAN MATHS
New research highlights the links between learning a foreign language and learning to code

ew research has shown that accounted for. Numeracy was found to

N aptitude for learning foreign explain variance in learning rate only.
languages is a stronger predictor of In many schools, a certain level of maths
learning to program than basic maths attainment is often a prerequisite for
knowledge. The study, led by Chantel studying computer science. While Prat’s
Prat at the University of Washington study was carried out on adults, its results
and published in Scientific Reports, raise questions around this widespread
examined various aspects of adults’ practice. “I don’t have a good reason to think
cognitive abilities, including language the predictive utility of language, numeracy,
and maths skills, while they learned general cognitive abilities or neural metrics n Participants learnt Python while wearing a specialised headset
that measures electrical activity in the brain. Image credit: Justin
to code in Python for the first time. would change in young learners,” she says. Abernethy/U. of Washington

The researchers tracked how quickly “Learning programming languages in and

participants proceeded through of itself doesn’t require knowledge of linear or the data to suggest that complex math
programming tutorials, how accurate algebra,” she explains. But maths remains isn’t required for computer science. In fact, I
their code was, and how well they an important skill in many applications of strongly suspect that there are many aspects
understood the programming language. programming. “I think the thing to keep in of engineering that do rely on math!”
Language aptitude was found to explain mind is that programming is still primarily The researchers chose to carry out
variance in all three measures of skill, taught in computer science and engineering the study in Python because of its close
once other areas of cognitive ability were departments. I don’t have the expertise resemblance to English structures such as
paragraph indentation and its use of real
words rather than symbols for functions.
Prat believes, however, that her results will
carry over to more complex languages,
such as Java. “I believe we’ll find that the
programming languages that are anecdotally
perceived as being intuitive or easy to learn
or read may be those that most resemble
human language structures, and learning
those languages may be most strongly
predicted by language aptitude. However,
all programming languages emulate human
communication structures and so I do think
the relationship between language aptitude
and programming aptitude will show up
with different languages and in different
learning environments. It’s a question
n Cognition and language aptitude are greater predictors of learning than numeracy. Image credit: Prat et al./Scientific Reports we want to pursue empirically.”

helloworld.cc 7
NEWS FEATURE

YOUNG CREATORS WOW

JUDGES AT COOLEST
PROJECTS USA
150 attendees showcased their inventions in Santa Ana, California

Kevin Johnson

oolest Projects USA returned on The judges were faced with difficult for Deforestation Station, a game designed
C Saturday 7 March for its third decisions in selecting the winners of each to help people better understand
celebration of the USA’s creative and category. Kausthubh, Maria, Sarah and Eric deforestation, as well as provide solutions.
innovative young creators. Held at won the Hardware award with Vest Buddy, The Overall Winner of Coolest Projects
Discovery Cube Orange County, a science 2nd Generation — a wearable technology USA 2020 was awarded to Adarsh, for his
centre in Santa Ana, California, the project that helps non-verbal young people project Vital Signs. Adarsh got involved with
technology fair allowed young innovators communicate. The group also attended digital making when he started attending a
to showcase their work. It saw young Coolest Projects last year and impressed local makerspace with his big brother. He
people attend from Northern California, the judges by showcasing how they created a system for monitoring vital signs
Texas, Lousianna, and Georgia. incorporated feedback into their project. The to diagnose illness without touching the
Anyone under the age of 18 could enter winner of the Visual Programming category patient. It also provides limitless possibilities
their creations into one of the fair’s was Nathan, who devised Memory from preventing Sudden Infant Death
categories of Hardware, Web, Mobile Apps, Assistant, a system to remind older people Syndrome to diagnosing viruses.
Visual Programming and Games. Last year’s to stay hydrated and take medication
winner of the Mobile Apps award, Ashley when needed.
ANYONE UNDER
Chu, featured as a special guest speaker.
She gave advice for anyone feeling nervous,
The Girls Who Code Club from Maywood,
California, showcased several projects, “ 18 COULD ENTER
while using what she called the ‘power inspired by things that are important to
pose’ accompanied by a wiggle dance to them. Kayla, Carola, Cassandra, Natalia and THEIR CREATIONS
shake any nerves away. Sibone from the club won the Games award

n Members of the Girls Who Code club from Maywood,

California, won the Games award.

8 helloworld.cc
 All images © Mike Villa

n Vest Buddy is a wearable technology project that

helps non-verbal young people communicate.

n Young creator Nathan won the award for Visual Programming

for his project Memory Assistant, which helps older people
stay hydrated and gives reminders to take medication.

helloworld.cc 9
NEWS FEATURE

n The Megaprocessor is a major exhibit

at the Centre for Computing History

MUSEUM SEEKS NEW TRUSTEE

The Centre for Computing History in Cambridge, England, has
an opportunity for educators to guide its management

Liz Upton

he Centre for Computing History The role of a trustee

T is seeking at least one new trustee As the centre is a registered charity, trustees
for its Board. They welcome applications have a series of statutory responsibilities and
from people with a wide range of skills and also act as ambassadors for the museum.
experience, and are particularly seeking a They steer the organisation in pursuit of
trustee with a background in education. its charitable aims. Trustees are expected
The Centre for Computing History is to use their specific skills, knowledge, and
a young museum based in Cambridge, experience to help the board of trustees to
England, with a growing visitor base and reach sound decisions for the benefit of all
an established and successful learning stakeholders. The role is not remunerated,
programme. It offers a hands-on learning but reasonable expenses can be met.
experience wrapped up in a historical Trustees are appointed for a term of four
context. Visitors can touch, program, and years, extendable for up to three terms. n There’s lots to discover for all ages at the museum

game on original hardware and learn about Board members will need to attend trustee
the people who created the technology board meetings, normally held six times per are required to attend an induction process
should be we rely on today. The Centre year in Cambridge, and occasionally promote at the museum.
welcomes visits from all schools, FE Colleges, the museum at events and meetings with Further details on the role and
and universities, and any organisations key people. They will generate and process how to apply are available at
supporting the education of young people. key documents and papers. New trustees computinghistory.org.uk/news

10 helloworld.cc
 NEWS FEATURE

n C lare Parry-Jones, UK Enterprise Sector Director at Computacenter,

presents Aoibheann and Avye with their awards

GIRLS RECOGNISED FOR

INSPIRING LEADERSHIP
Avye Couloute and Aoibheann Mangan were recognised for their work in shaping
the future of the technology industry at the FDM everywoman awards

wo young leaders have been there are loads of people who want to and asking whether her daughter was
T recognised for their efforts share their knowledge and ideas with going to come along. Her friend said ‘Ah,
in inspiring more girls into careers in you. They also allowed me to craft, code, no, she thinks it’s for boys…’ A few weeks
tech. Avye Couloute and Aoibheann design and to discover new things,” she later and she was first in the door to come
Mangan were announced joint winners writes on her blog, 10tonolimit.com. along. She had heard from her friends that it
of the FDM everywoman ‘One to Watch’ Avye’s work was also recognised was a girl doing the workshops and she felt
award at the 2020 FDM everywoman in last year when she won The Diana inspired to come! That was lovely to hear.”
Technology Awards earlier this month. Memorial Award. The Diana Award is a Aoibheann has also been at the forefront
Avye Couloute, 12, attends Surbiton charity founded in memory of Princess of a campaign for broadband equality in
High School in London. She creates Diana and her belief that young people Ireland. “In Ireland there is a large divide in
opportunities for girls, in particular, to have the power to change the world. areas where you can have broadband and
explore coding and robotics, through her Aoibheann, 13, is a student at Mount St. areas where you have dial-up or no internet
Girls Into Coding events. She has fundraised Michael Secondary School in Claremorris, at all,” she says. “I live in rural Ireland where
and secured support to provide girls with Ireland. She runs a CoderDojo at her school, this problem is prevalent. I spent many hours
tech-themed books and physical computing where she mentors over 50 children weekly, sitting in my local car park in Tesco to get
kits. Avye is ambitious about the scope with 60 per cent of them being girls. WiFi to complete projects. I found this very
of her work: “It’s a big goal but this year I “I began coding when I was 8,” Aoibheann unfair and I’m mindful in times like this with
want to reach one thousand girls through explains. “I fell in love with it from day the coronavirus worries that a lot of schools
my events.” Girls Into Coding is currently one. Mum tried to find a CoderDojo for us may have to use remote classroom learning
sponsored by Microsoft but Avye and to attend but there was none in our area in the coming weeks, that students may
her mum, Helene, are currently looking at the time so she set one up. That was be faced with being unable to complete
for new partners to expand their work. something that inspired me when I went schoolwork due to lack of broadband. It’s
Avye started to become interested into first year in my secondary school, to important everyone has a fair chance.
in computing at the age of seven, after try and set up a lunchtime coding club.” I’d love to be able to remote work in this
attending a workshop with her mum. “When I first started a local Dojo, I can beautiful part of Ireland some day and not
“I really enjoy going to tech events as remember Mum meeting one of her friends be faced with moving away to work!”

helloworld.cc 11
INTERVIEW
OPINION

DAVID F
or someone who is arguably the most famous
living statistician, David Spiegelhalter is
surprisingly scathing of his field. “When I

SPIEGELHALTER
complain about statistics, I just think of old-fashioned
education, where it appears like, ‘Is it a chi-squared? Is it
a t-test? Is it one-sample or two-sample? Is it a
regression? Which test do I slap on and get some answer

ON TEMPERING THE coming out of the program?’ And that’s dire, it really is.”
Spiegelhalter is letting rip at his field while sitting in the

HYPE AROUND AI central court of the Centre for Mathematical Science, where
he is the Chair of the Winton Centre for Risk and Evidence
Communication. The Centre’s buildings are beautiful, with
sweeping curved walls and low, grass-covered roofs. “We
hope the layout of the buildings and the way they work
The statistician meets Sian Williams Page will actively encourage fresh dialogue and discovery,” the
to discuss shoddy statistics, overblown award-winning architects explained to The Guardian in
2003. Though perhaps architecture can only go so far:
algorithms, and a golden age for just inside the entrance, there is a sign urging staff and
artificial intelligence students to smile at one another, and to try saying hello.

12 helloworld.cc
 Spiegelhalter doesn’t seem to need a reminder to smile play games. But these are deeply restricted classes of
and talk. Over the course of his career he has shifted from problems: you need no background knowledge, you just
working on serious maths to improving how evidence is need a huge amount of information on past cases. The idea
communicated to the public. In his latest book, The Art that ‘Oh, because a program can learn Go [in 2016, Google
of Statistics, he applies what he calls ‘simple numeracy’ DeepMind’s AI program AlphaGo beat world master Lee
to messy, real-world problems. The questions he tackles Sedol at the board game], therefore you can let it free on
include ‘Do speed cameras reduce accidents?’, ‘What’s medical records and it can tell you what’s going to happen
the cancer risk from eating bacon sandwiches?’, and to a patient,’ is complete nonsense, just nonsense! And
‘Does going to university increase the risk of getting a there’s so much overhype and over-claiming being made.”
brain tumour?’ (The answer to the final question is a If anyone is qualified to raise an eyebrow at the current
resounding no.) enthusiasm for machine learning and artificial intelligence,
He hopes the book provides a grounding in it’s Spiegelhalter. Some 35 years ago, he and a colleague
statistical principles for new graduates being lured into applied probability theory to artificial intelligence with
careers in data science — a multidisciplinary field that significant consequences: the work has been cited
combines statistics with machine learning, design, and thousands of times by other researchers and underpins
communication. He welcomes the advent of the subject many machine learning algorithms used today. He
and the facelift it has given statistics. “I think the ability to brightens when recalling the period. “When you look back
deal with data critically and on your career, you think there
to realise its strengths and are certain times which have
limitations is the most important
DATA SCIENCE IS AN
“
been a bit of a slog, and other
skill in the future world and it’s times it was really exciting and
an extremely marketable skill as ESSENTIAL SKILL IN energising. The eighties were

TODAY’S WORLD
Jonathan Howard/Raspberry Pi Foundation

well,” he explains. really exciting.”

But in the shift from statistics The challenge educators
to data science, has one bag of face to convey a balance
tools simply been replaced by of enthusiasm for machine
another? Are machine learning algorithms being applied learning and artificial intelligence with a healthy dose
with little care for their inner workings, much like the of scepticism to their students is one Spiegelhalter is
statistical tests of Spiegelhalter’s early years? keenly interested in. Having previously advised England’s
“Machine learning is even worse! Because you can get Department of Education on the syllabus for GCSE
a set of data, and then you bang in a logistic regression, Mathematics, he is now a member of a Royal Society panel
and classification trees, random forests, neural networks, that provides recommendations on data science in schools.
small vector machines and off you go, bam, bam, bam. You In 2018, he co-authored a report that argued for a ‘more
apply them all and you’ve got no idea what they are doing detailed, updated, and coherent computing curriculum,’
and out they come with some error rate and you think ‘Oh, with recommendations for introducing content on machine
that one’s best.’ And this is appalling, absolutely dreadful, learning and computational modelling and strengthening
because you’ve got no idea,” he says. content on data representation.
“The robustness and explainability and fairness and “Data science certainly does not belong in maths. I don’t
transparency are absolutely vital when it comes to this even really believe statistics belongs in maths. Data science
work. Frankly, unless you’ve got a pretty good idea of in all its manifestations — sorting data, getting conclusions
what it’s doing you’ve got no guarantee whatsoever that if from data visualisation, communication — is an essential
you move into a slightly different domain that it won’t just skill in the world and it’s one that should be part of schools’
all fall apart. You slightly change the inputs, move into a education, but I don’t know where it should be. It has a
different context, and you’ve got no idea why, suddenly, it broad spectrum of literacy that can be applied across all
doesn’t work.” areas — geography, the humanities, computer science —
Spiegelhalter’s is not just sceptical of how businesses nobody owns it, and it’s wonderful. I wish I could say ‘You
apply algorithms, but also of the ambitious and broad must do it this way,’ but I really don’t know. But, I suppose
claims made about the technology. “You have to recognise it’s quite right you should go into something not knowing
there are certain problems and challenges that machine the answers.”
learning and AI have been absolutely fantastic at: well-
defined problems in terms of optimisation. I use Google The Art of Statistics is published by Pelican and is available
Maps — it’s a fantastic optimisation algorithm, there’s to purchase in paperback now: helloworld.cc/stats.
fantastic image recognition, there’s an amazing ability to

helloworld.cc 13
COLUMN
OPINION

THE INCLUSIVE
COMPUTING CLASSROOM
Catherine Elliott launches her new column with a reflection on belonging

I
am delighted to bring you a regular column on all included in lessons. (Obviously some students with severe
things related to special educational needs and learning difficulties or physical disabilities will still require
disabilities (SEND) and Computing. Over the specific interventions.)
coming issues I will go into more depth about the strategies Consider how you measure success in your classroom.
and resources that teachers can use to ensure their SEND Some students may struggle to create a working program
learners are accessing the subject in a meaningful way. in Python, Scratch, or on the Bee-Bot. This is where the
Why is this important? Many of our young people PRIMM model is really useful. If you haven’t heard of PRIMM
with SEND use technology to access information and yet, then head to suesentance.net/primm-project to find
entertainment, just like their peers. Some of them will rely out more. Essentially it provides multiple access points to
on computers to communicate and to access their learning interact with code. A student can read existing code and
in school. We need to ensure that they are taught how to do discuss what they can identify and predict what they think
this safely, responsibly, and effectively. Our students with it will do. Another student can take a working program,
SEND also need to have the same economic opportunities and make small adaptations to it, for example change the
presented by studying the subject. Furthermore, background, alter the values in a variable, change
we need them to be represented in our future the contents of a string. A young person
workforce – the digital devices and products with SEND can thus learn about the same
of the future will only be as diverse as the computer science concepts as their peers,
people who design and test them. without the fear of failure, or the demand
I have recently been reading a great on working memory and recall that writing
deal around inclusive design, and this a program from first principles involves.
is one strategy for supporting students We could also offer a range of other
with SEND. You can read more about this programming-related activities during
at microsoft.com/design/inclusive/ and lessons, for example using unplugged
mismatch.design. “Inclusive design doesn’t activities to teach key concepts, spending
mean you’re designing one thing for all time designing programs in groups, or
people. You’re designing a diversity of Catherine is the SEND lead for working with physical computing devices
the Sheffield eLearning Service
ways to participate so that everyone has that have a multimodal output (light/
(sheffieldclc.net), and she has spent
a sense of belonging.” Susan Goltsman, sound/movement). More on these in a
the last five years working on ways
(quoted in Mismatch by Kat Holmes, future column!
to make computing accessible for all
2018). This sense of belonging is at the learners. She is a member of the CAS So take time to reflect on what happens
heart of inclusion. If, as educators, we can Include working group, and co-leader in your own classroom. How diverse are
provide multiple ways of engaging with of the Sheffield and South Yorkshire the methods of interaction in your lessons?
the learning in the classroom, then we Secondary CAS Community. Are all students included in
can ensure more of our young people are @catherinelliott the learning?

14 helloworld.cc
 SNAPSHOTS FROM
THE HISTORY OF AI
George Boukeas introduces fascinating
stories to share with your students

he story of artificial intelligence to think. Instead, he suggested a game as in terms of any internal processes. Humans
T (AI) is a story about humans a proxy for answering the question: the are still the yardstick for intelligence, but
trying to understand what makes them imitation game. there is no requirement that a machine
human. Some of the episodes in this In modern terms, you can imagine should think the way humans do, as long as
story are fascinating. These could help a human interrogator chatting online it behaves in a way that suggests some sort
learners catch a glimpse of what this field with another human and a machine. If of thinking to humans.
is about and, with luck, compel them to the interrogator does not successfully In his essay, Turing also discusses learning
investigate further. determine which of the other two is the machines. Instead of building highly complex
human and which is the machine, then programs that would prescribe every aspect
The imitation game the question has been answered: this is a of a machine’s behaviour, we could build
In 1950, Alan Turing published a machine that can think. simpler programs that would prescribe
philosophical essay titled Computing This imitation game is now a, fiercely mechanisms for learning, and then train
Machinery and Intelligence, which started debated, benchmark of artificial intelligence the machine to learn the desired behaviour.
with the words: “I propose to consider the called the Turing test. Notice the shift in Turing’s text provides an excellent metaphor
question: Can machines think?” Yet Turing focus that Turing suggests: thinking is to be that could be used in class to describe the
did not attempt to define what it means identified in terms of external behaviour, not essence of machine learning: “Instead of
trying to produce a programme to simulate
the adult mind, why not rather try to
produce one which simulates the child’s? If
this were then subjected to an appropriate
course of education one would obtain
the adult brain. We have thus divided
our problem into two parts: the child-
programme and the education process.”
It is remarkable how Turing even
describes approaches that have since been
evolved into established machine learning
methods: evolution (genetic algorithms),
punishments and rewards (reinforcement
learning), randomness (Monte Carlo tree
search). He even forecasts the main issue
with some forms of machine learning:
opacity. “An important feature of a learning
machine is that its teacher will often be very
largely ignorant of quite what is going on
inside, although he may still be able to some
extent to predict his pupil’s behaviour.”

helloworld.cc 15
FEATURE

The evolution of a definition algorithm to be implemented into a program a solution “will act as a wedge in attacking
The term ‘artificial intelligence’ was and Turing famously played a few games other problems of a similar nature and
coined in 1956, at an event called the with human opponents executing the of greater significance”. Indeed, search
Dartmouth workshop. It was a gathering algorithm by hand. Claude Shannon wrote was the driving force behind many of the
of the field’s founders, researchers who Programming a Computer for Playing landmark achievements in the field: making
would later have a huge impact, including Chess, in 1950, where he laid the foundation plans and schedules, proving theorems
John McCarthy, Claude Shannon, Marvin for many of the search techniques that would with logic, solving algebraic problems,
Minsky, Herbert Simon, Allen Newell, later be applied in games. making inductions, and so on.
Arthur Samuel, Ray Solomonoff, and The main approach for playing many
W.S. McCulloch. of these games is search: systematically Deep Blue
In 1996, the chess world champion Garry
Kasparov played against Deep Blue, a

“ INTELLIGENCE IS THE QUALITY THAT

ENABLES AN ENTITY TO FUNCTION
purpose-built IBM computer. Deep Blue
became the first chess machine to ever
win a game and, a year later, a match
APPROPRIATELY AND WITH FORESIGHT against a world champion under regular
time controls. Kasparov’s defeat made the
headlines and is considered a milestone in
The simple and ambitious definition generating and evaluating positions the history of AI.
for artificial intelligence, included in the and moves. That may sound trivial for a The main driving force behind Deep Blue
proposal for the workshop, is illuminating: computer but the number of combinations was search: it was able to generate and
‘making a machine behave in ways that in non-trivial problems quickly explodes evaluate 200 million positions per second.
would be called intelligent if a human exponentially and a brute-force The evaluation function was handcrafted
were so behaving’. These pioneers enumeration of all the possible outcomes is by human experts and the only form of
were making the assumption that ‘every impossible. Shannon estimated the number learning was the system’s ability to fine-
aspect of learning or any other feature of different chess games to 10120. In these tune some of its parameters.
of intelligence can in principle be so vast search spaces, a lot of thought needs Modern chess-playing programs need
precisely described that a machine can to go into evaluating search states to guide to evaluate far fewer positions and do
be made to simulate it’. This assumption the search effort and prune non-promising not require specialised hardware to vastly
turned out to be patently false and led search paths. outperform human players. The last known
to unrealistic expectations and forecasts. In his paper, Shannon discussed the win by a human against a top chess-
Fifty years later, McCarthy himself stated value of research in games, explaining that playing machine was in 2005.
that ‘it was harder than we thought’.
Modern definitions of intelligence are of
distinctly different flavour than the original
one: ‘Intelligence is the quality that enables
an entity to function appropriately and
with foresight in its environment’ (Nilsson).
Some even speak of rationality, rather than
intelligence: ‘doing the right thing, given
what it knows’ (Russell and Norvig).

Playing games: search

A lot of research in artificial intelligence
focused on games. Over the years,
programs for checkers, backgammon,
and many other games have reached
competence levels that surpassed the
best human players.
However, chess was the most prominent
game, right from the start. Alan Turing and
David Champernowne developed a basic
algorithm called Turochamp for playing
chess back in 1948. It took years for that

16 helloworld.cc
 Deep Blue’s victory was part of an that uses multiple different techniques the weights. The function computed
impressive string of achievements, and to answer questions posed in natural is a composition of simpler functions –
yet there were problems that seemed language. In order to answer a question, represented by individual neurons. To
elementary and yet proved extremely hard Watson generates multiple hypotheses and build a neural network boils down to how
to tackle. This was eloquently summarised seeks to support them by drawing evidence these simpler functions are organised and
in Don Knuth’s remark: “AI has by now from a body of sources. In other words, composed (the network ‘topology’), also
succeeded in doing essentially everything Watson is able to provide justification for taking the weights into account. A neural
that requires ‘thinking’ but has failed to its answers. There are many areas where network can ‘learn’ in the sense that its
do most of what people and animals do Watson is now being applied, the most weights can be modified, swaying the
‘without thinking’ – that, somehow, is much prominent being in assisting doctors with output in more desirable directions.
harder!” It is only very recently that artificial diagnosis and suggested treatment. Behind every recent breakthrough in
intelligence has made breakthroughs in the Watson’s level of complexity is artificial intelligence, you will find a neural
latter class of problems, such as image and astonishing and it would be impossible to network. Teaching a neural network of
speech recognition, and this is the main develop such a system without some form sufficient complexity requires a significant
reason it has become so prominent. of learning. Echoing Turing’s comments amount of training instances and
about learning machines, Grady Booch, computational power. Even though neural
Watson who was involved in building Watson,
In 2011, a computer system built by IBM remarked that “building a cognitive system
and named after the company’s founder, [like Watson] is fundamentally different GEORGE
competed against two human champions than building a traditional software- BOUKEAS
in the game of Jeopardy! The highly intensive system of the past. We don’t George is a Learning
publicised match resulted in an impressive program them. We teach them.” Manager for
win for Watson, in a context that would secondary education
traditionally be considered extremely hard Neural networks at the Raspberry Pi
for a computer to tackle. A neural network receives input values Foundation
Watson is not really a computer system and computes output values, which are (@pythonies).
for playing Jeopardy!, though. It is a system influenced by a set of parameters called

helloworld.cc 17
FEATURE

GO WAS
“ CONSIDERED
THE HOLY
GRAIL OF
GAME AI

networks have been around for decades,

it has only been in the last few years that
their potential has been realised, as the
amount of available training data has
skyrocketed and computing power, along n Neural networks are behind nearly all AI technology
with dedicated hardware, has become
more readily available. that reliably evaluate positions in order to games or uses self-play, in order to learn
guide the search. Researchers estimated how to evaluate positions and moves.
AlphaGo that it might take decades for machines to It uses neural networks to compute its
Go is an ancient strategy game for two beat humans at Go, which was considered evaluation functions and modify them
players, who take turns placing black and to be the holy grail of game AI. while learning. It searches through the
white pieces (stones) on a 19x19 board. Enter AlphaGo, a computer program by vast space of possible positions by
It is a notoriously difficult game for DeepMind. In 2017, AlphaGo beat Ke Jie, taking random samples, instead of
computers. The number of possible board the world’s top-ranked player at the time, searching systematically.
positions is estimated at an astronomical following victories over other high-ranking This is such a promising generic
10170. Traditional search techniques in such professional players. approach that AlphaZero, a generalised
a vast space of possibilities is pointless and AlphaGo combines previously known version of the program, used only self-play
it has proved very hard to develop functions methods in a novel way. It studies human to achieve a superhuman level of play in
the games of chess, Shogi and Go within
24 hours. This is a step closer to Turing’s
vision of a blank slate child-programme,
endowed with the ability to learn.

FURTHER READING
Artificial Intelligence,
Michael Wooldridge. Ladybird, 2019.

Machine Learning for Humans,

Vishal Maini and Samer Sabri, 2017.

The Quest for Artificial Intelligence:

A History of Ideas and Achievements.
Nils J. Nilsson, Cambridge University Press, 2010.

Machines Who Think,

Pamela McCorduck, A K Peters, 2004.

18 helloworld.cc
 POPBOTS OPEN AI TO THE
YOUNGEST LEARNERS
From Siri to gaming, artificial intelligence is all around us.
Ben Hall explores an innovative project run by MIT that makes
this technology accessible to preschoolers

ery often, the most vivid examples their way around a tablet, and merely Research-led
V of how quickly children learn and exposing them to it will not introduce new PopBots are still in the early stages of
progress can be seen in the youngest age concepts. How can we develop children’s development and not yet widely available.
groups. A great example of this is how computing knowledge so it will encourage Despite this, there is already a growing body
children use construction-based resources them to be more than just passive users? of research investigating the inclusion of AI
such as wooden blocks, train tracks or in the curriculum at an early age. Williams
LEGO. Often, children experience these for AI literacy would like to see PopBots developed as an
the first time when they enter the reception Could AI provide an answer? A team at MIT open-source platform that students could
classroom. Initially, they will use these has addressed how to make AI accessible build from classroom materials. Research
resources in quite a free, unstructured way. by developing PopBots, small robots aimed supports further development.
Over time, with guidance and support, at introducing AI to young learners. They A recent paper by Williams and colleagues
they construct increasingly complex and have also developed easy-to-use resources investigates how AI can influence young
imaginative models, often linked to other to complement the hardware. children’s perceptions of robots (helloworld.
areas of learning. The models from the end One such project explores using machine cc/robots). They found that perceptions of
of the year are unrecognisable from those learning, where a robot can be programmed robots can be shaped at an early age, so
constructed at the beginning. to sort healthy or unhealthy foods. Children for children to be AI-literate, their earliest
Can the same be said of children’s begin by classifying foods for the robot, but experiences should be meaningful and
first encounters with technology in the soon realise that it would take too long to informed. As it is an emerging technology, the
classroom? Many schools will provide access do this for every single food type. Through research is at an early stage. There is no doubt
to tablets or computers as part of their early some supervised machine learning, children that our world will be increasingly shaped by
years provision, but how much of that is can quickly train PopBots to classify foods, AI, so helping children develop a conceptual
about consuming the technology, and do also developing their own understanding of understanding at an early age needs to be at
children relate this technology to real-world healthy eating. Through this simple activity, the forefront of curriculum development.
applications? In my experience, simply children become AI-literate creators —
having the devices available is not enough turning a passive device into something that
— most children start school able to navigate makes informed, intelligent decisions. BEN HALL
Randi Williams, a PhD student at MIT, who Ben Hall is a
worked on the PopBots project, provides Learning Manager
at Raspberry Pi,
WHAT IS A POPBOT? her view of introducing children to AI:
‘Children’s views of themselves relative to where he develops
the technology change. Their views of how resources for the
PopBots are constructed using LEGO and use a much they can participate in technological NCCE. He is a CAS
mobile phone with additional LEGO or Arduino invention change. I love the fact that early Master Teacher
peripherals. Users interact with them via a AI education makes children feel more and a Raspberry Pi Certified Educator
programming interface on a tablet or computer. curious about their world and empowered (@hengehall).
to change it.’

helloworld.cc 19
FEATURE

NATURAL LANGUAGE
PROCESSING WITH SNAP!
Students can write programs to solve puzzles like ‘Hat is to
head as glove is to X’ with the block-based software

oday’s AI programs can generate What might school students do with vector arithmetic blocks. More advanced
T recommendations, answer word embeddings? Can they be made users can instead take advantage of
questions, summarise text, analyse accessible to non-experts? What might Snap!’s support of higher-order functions
sentiment, and perform translations. To do they learn in the process? to use list mapping blocks to perform
this, they rely on a technique that treats We attempted to answer these the vector operations. Students may
words as long lists of numbers. These AI questions by building a Snap! library that have encountered linear algebra in their
programs are based on neural networks. contains 20,000 word embeddings in 15 studies (students in the UK, for example,
Neural networks cannot deal directly with languages. Using a block that reports a
textual data – input text must be converted list of 300 numbers for any of the known
to numbers, and output numbers are words, students can create programs that KEN KAHN
often converted back to text. The list or search for similar words, find words that Ken is a researcher
vector of numbers associated with each are the average of other words, explore in the Department
word can be understood as coordinates. cultural biases, and solve word analogy of Education at the
The technique is called ‘word embedding’ problems. These programs can work in a University of Oxford.
because it places each word somewhere in single language or rely upon the alignment In the 1970s he
a high-dimensional space. This is a space of the word embedding spaces of different was a member of
that is impossible to visualise and very languages to perform rough translations. Seymour Papert’s
hard to understand. But putting words into To compute with word embeddings, Logo Group at the
this space enables AI programs to do some we need to perform vector arithmetic. MIT AI Lab.
very impressive things. This can be accomplished by providing

20 helloworld.cc
are expected to learn to apply addition
and subtraction of vectors and multiply
vectors by a scalar between the ages LEARNING OUTCOMES
of 11 and 14). They may be surprised
to discover how useful it is in dealing Trials of these resources are being performed by our colleagues at Beijing Normal University.
intelligently with words, too. We plan to investigate how the resources can relate to:
n Linear algebra (appreciation, creative uses, and practice)
Word embedding Snap! blocks n Higher-order programming (understanding and appreciation)
A surprisingly wide range of applications n emantics of words (synonyms, antonyms, word senses, and so on)
S
is possible with only two blocks: a n eometry (high-dimensional spaces, correspondences with vector arithmetic)
G
block that, given a word, reports a list nA
 I (natural language processing and machine learning)
of numbers; and a block that, when nC
 omputational thinking in a context rarely considered
presented with a list of numbers, reports
which words are closest. We also provide
blocks reporting all known words, and reached from helloworld.cc/craft. The AI words such as ‘man’ and ‘boy’, and
blocks for helping to visualise the blocks are described in an article in Hello compare these to the distances between
locations of words. Some functionality World issue 5. the word ‘president’ and female words
such as ‘woman’ and ‘girl’, we find that
‘president’ is closer to the male words

“ WORD EMBEDDING PLACES WORDS IN A HIGH-

DIMENSIONAL SPACE THAT IS IMPOSSIBLE TO
than the female words. The opposite is
true for the word ‘maid’.

VISUALISE AND HARD TO UNDERSTAND Solving word analogy problems

An amazing aspect of word embeddings
is the way they can be used to solve word
is provided with both simple and full- Finding biases in AI analogy problems. For example, we can
featured versions of blocks. Determining how close words are to each solve the problem ‘Cat is to kitten as dog
The blocks are available as a Snap! other relies on data from AI programs is to X.’ We can solve for X as (cat – kitten)
library, as a Snap! project that illustrates trained on all Wikipedia articles, and a + dog. In (cat – kitten), we are calculating
typical uses, and as an interactive large web crawl. These pages reflect the youth of the word. If we then add that
guide. These webpages are part of a societal biases. We can create Snap! to ‘dog’, we get a ‘young dog’ – or in other
larger library of AI Snap! blocks that programs to uncover these biases. For words, a puppy!
was produced as part of the EU-funded example, if we look at the distances Word analogy problem-solving using
eCraft2Learn project. All this can be between the word ‘president’ and male word embeddings is not limited to the
meaning of words. Word embeddings
also encode how words relate to one
another. So, for example, they can solve
‘Fast is to X as slow is to slower.’
Biases can also be found in word
analogy solutions. Soon after researchers
discovered how word embeddings could
be used to solve analogy problems, it was

n Technology such as Amazon’s Alexa is

based on natural language processing

helloworld.cc 21
FEATURE

pointed out that their database declared training with billions of sentences, a part to that word, while making sure never
that ‘Man is to doctor as woman is to of the neural network was extracted to to repeat the same word. Use this to
nurse.’ In an effort to eliminate this and create the word embedding databases. repeatedly change random words one at
other similar biases, our Snap! blocks are a time in a famous poem or text (such as
based upon a newer database. Possible projects ‘roses are red and violets are blue’).
Our Snap! blocks used the databases Some suggestions for projects using n Make word games using word
found at fasttext.cc, an improvement these word embedding blocks are: embeddings. For example, try something
over the original word embedding work n Use word embeddings to explore the like Semantris (helloworld.cc/semantris),
called word2vec (helloworld.cc/word). similarity of sentences. One idea is to a semantic version of Tetris. Working
Both projects involved training a machine average all the words in the sentence. towards a bilingual version of Semantris
learning system to predict which words This is called the ‘bag of words might be a good idea.
had been removed from sentences. technique’ as it ignores the order of the n Create a program that searches for new
One or two words per sentence were words, just as if they were put in a bag. word analogies. Hint: If A is to B as C is
randomly replaced by a blank token and n Find a chain of similar words by finding to D, the distance A–B ‘is close to’ the
the program was trained to compute what the nearest word to the starting word. distance C–D.
were the most likely missing words. After Then repeatedly find the nearest word n Search for biases that arise from the
way people write about things. See if
any biases you come across also apply
in other languages.
n Use word embeddings as a way to
input textual data into the deep neural
networks. The deep learning models can
be created with other Snap! AI blocks.
This opens up a wide range of natural
language processing possibilities.

Availability
All of these resources work in modern
browsers. They are free and open-source.
They can be run on a local web server without
any internet connection. Visit helloworld.cc/
n Snap! blocks answer questions about words by doing arithmetic craft for more information.

22 helloworld.cc
 MACHINE LEARNING IN
THE PRIMARY CLASSROOM
Nicholas Hughes explains why it’s important every child explores the
technology for themselves, and how he’s broaching the topic with his classes

achine learning is all around us, in through ideas such as making their own
M our homes, on the websites we use, version of a smart assistant.
and in our pockets. More parts of our lives I started simply and used Dale’s ‘Make MACHINE LEARNING
are being affected by it, from suggestions Me Happy’ resources to create a program
on streaming services and shopping sites that would recognise nice and nasty RESOURCES
to automatic tagging of our photos. How phrases (see pages 80–82). This linked
does all this happen? How do we know well to some work we had just been If you want to learn more about machine learning
that the answers are right? What data is doing: programming a face to react to such and investigate some resources, I recommend:
being used to make suggestions? We all phrases. In this activity, a face reacted to n Dale Lane’s (@dalelane) Machine Learning for
need to have an awareness of data quality specific phrases that the class offered up. I Kids – machinelearningforkids.co.uk
and bias. I believe all children and adults had asked the class how we could make the n Code.org AI for Oceans – code.org/oceans
should have an understanding of what face react to any phrase, and they agreed – great videos that are easy to understand
machine learning is and how it works. that it was impossible to write everything, n Quick Draw – quickdraw.withgoogle.com

After hearing Dale Lane talk so this was a great opportunity to introduce – can Google recognise what you have drawn?
about the resources on his website, machine learning as a solution. n Teachable Machine – teachablemachine.

machinelearningforkids.co.uk, I was We spent most of the lesson creating the withgoogle.com – train a computer to
keen to get my students exploring model. We started by discussing that we recognise your own images, sounds and poses
the technology. I was struck by how needed to collect data for nice and nasty
accessible the activities are and how they things and place them in the right buckets.
can be used to explore some of the wider They learnt that machine learning models It provided an excellent springboard
implications of machine learning in class. don’t always get it right, so more data is to discuss where else machine learning
It is also a fun and exciting way to engage required to retrain the model. models exist, e.g. spam filters, photo
children with computer science concepts, tagging, smart speakers, etc.
The Machine Learning for Kids website
contains prebuilt Scratch projects to
NICHOLAS demonstrate how the machine learning
HUGHES model can be used. There are also activities
Nicholas is a from the other resources mentioned above.
Computing These provide children with opportunities
teacher at to explore and start thinking about the
Latymer Prep implications for machine learning use.
School and an Machine learning is all around and
Educational providing opportunities for children to
Consultant at 3BM Education. He is a explore what it is and to make their own
CAS Master Teacher and a Raspberry Pi- projects is an important step in ensuring
Certified Educator (@duck_star). children have a greater knowledge and
n Hughes’s students enjoyed using the activities available
on the Machine Learning for Kids website understanding of this technology.

helloworld.cc 23
FEATURE

MACHINE LEARNING COURSES

TO DIVERSIFY COMPUTING
Tayler Kirk explains how new courses from Apps for Good could encourage more
girls and under-represented groups of students to choose to study computing

ith an already brimming Computing And this inspiration is crucial. Although Depth and breadth
W curriculum, some teachers may be figures from the Joint Council for At Apps for Good, we recognised how
hesitant about the prospect of teaching Qualifications (JCQ) have shown a rise of powerful machine learning courses could
machine learning. The technology may seem 7.2% in GCSE Computing qualifications, just be in engaging students who might not
too daunting for many educators to broach 21.4% of these were achieved by female otherwise have considered computing to be
with their students. The concept centres students in 2017. This figure is even worse a subject for them. Over the course of 2019,
on a computer rather than a programmer, at A level: only 9.8% of those completing a we have developed two courses in which
deciding and developing its own set of Computing course were girls. students work in teams to build and use a
instructions on how to sort, organise, If we can make teaching technology machine learning model to help solve a real-
classify, and use information. There’s a lot relevant to our students’ lives rather than world problem. Students begin by learning
of terminology that can be off-putting, and abstract and dry concepts, this trend can the core concepts of machine learning,
at its heart there are algorithms that may be be reversed. Interventions that highlight before going on to design and code their
beyond most people’s grasp. the relevance of science to students’ lives own models.
However, machine learning is have been shown to enhance their interest Our courses are designed with an
revolutionising our interactions with in the subject, particularly for students interdisciplinary approach that allows
technology, from running complex who are most at risk of being disengaged students to develop technical skills such
algorithms to utilising secure biometric from school. And a 2011 study of students as design and coding while also nurturing
data, and is now entering the educational in California showed that the strongest soft skills such as teamwork, problem-
landscape. By exposing students to predictor of girls’ interest in computing is solving, and presenting. While the course
machine learning, we unmask the everyday the extent to which they see its value. content doesn’t shy away from exploring
mysteries of how technology is influencing
and arguably improving our lives. By
teaching students how Spotify is able to
pick the perfect song, how Netflix knows
what you want to watch, and how chatbots
have averted the need to ever speak
to a human again, we can foster a real
enthusiasm for technology.

TAYLER
KIRK
Tayler is Education
Outreach and
Engagement Manager
at Apps for Good.
n Apps for Good have developed two machine
learning courses in which students work in teams

24 helloworld.cc
the inner workings of machine learning
algorithms (students work on activities such
as decision trees and k-means clustering,
among others), we believe that it is also
important to expose students to the ethical
considerations of such technology. So
our activities also encourage and facilitate
debates such as whose safety should a
driverless car prioritise, the implications
of storing biometric data, and whether
the police should be allowed to use facial
recognition technology. Opportunities to
present their thoughts and findings are
interspersed throughout the courses to
allow students to practise sharing their
ideas in an articulate and confident way.
Two students from Speyside in Scotland
recognised the immense power that n Isla and Phoebe from Speyside designed and began to prototype
a machine learning model (Ultra IR) that would use image
recognition to identify abnormalities in foetal ultrasound scans

MACHINE
“
chosen to begin delivering machine learning student workbooks, slides, lesson points,
via the programme. and primers so busy teachers can quickly
LEARNING IS NOW “We’re really excited to be delivering App access and deliver our materials completely
REVOLUTIONISING for Good’s machine learning course to five free of charge. All our materials have
year nine classes this year. We have been been designed by a former Computer
OUR INTERACTIONS delivering the app development course Science teacher and are mapped against
WITH TECHNOLOGY for the past eight years and wanted to the Computing curriculum, so they can
try something new. The machine learning be delivered in lesson time or as an
course is cutting edge and includes a good extracurricular activity. Each session has
machine learning could yield and how this amount of programming and computation a lesson primer so teachers don’t have to
could be used to assist in maternal care theory to fit in with our overall ambition spend time researching topics and can
for women in rural communities. Isla and of encouraging our girls to both enjoy and confidently deliver sessions on topics that
Phoebe designed and began to prototype develop their computing knowledge and they may not have previously come across. 
a machine learning model (Ultra IR) that skills. The fact that the students can speak Although all our courses, including our
would use image recognition to identify with experts in the field through the Apps ‘In a Day’ workshops, are designed to be
abnormalities in foetal ultrasound scans. for Good programme adds a real boost to as easy to deliver as possible, we know
As the machine was exposed to more the team’s engagement, too.” that some schools may not feel confident
examples of ultrasounds it would become to deliver a course. Since February 2018,
more attuned to pinpoint anomalies, Free to access members of Apps for Good’s staff have
particularly those that could be missed by We work with over 650 schools and been going into schools in challenging
human error. colleges across the UK and recognise circumstances across the UK to deliver ‘In
Their teacher, Marc McWhirter, said, “I that every teacher is in a unique position a Day’ workshops so that teachers can use
was thrilled to take part in the first year and that our Standard courses, which last the experience as a training opportunity
of Apps for Good’s machine learning between ten and twelve hours, may not and then feel comfortable to deliver the
course. I was initially worried about the be suited to everyone. For these teachers workshop to additional classes. Although
learning curve, but the team at Apps for who may be pressed for time, our ‘In a we are only a small team of ten, with two
Good presented the material and practical Day’ workshops allow them to explore members of staff facilitating sessions, we
activities in a way which made the topic far skills of app design or machine learning in are hoping that 2020 will see an expansion
more accessible than I had anticipated. The a condensed five-hour course that goes of our delivery capacity and we will be able
topic certainly fired up the imaginations through the same process of ideation, to reach more students than ever before
of my pupils, who are always keen to be design, build, and present, but within a and support educators in their delivery of
tech pioneers.” single school day.  our courses and workshops.
Lilli Hutchinson is a teacher who has Our courses are free of charge to access You can access all the Apps for Good
been delivering Apps for Good courses and published under a Creative Commons course materials for free after registering,
for over eight years. Her department has licence. They come fully resourced with for free, at appsforgood.org.

helloworld.cc 25
FEATURE

n The Get Creative With Data course was launched with

a preview event at Tapton School in June 2018

ARTIFICIAL INTELLIGENCE
IN THE CLASSROOM
Emily Dreimann shares some ideas for kick-starting discussions on the technology

here’s little doubt that artificial Robotics’ lifelike robot Sophia, for example, and interactions with the technology. Here
T intelligence has captured our has become a familiar sight on talk shows are just some of the angles from which this
collective imagination. TV series and films and has even starred in music videos. topic can be approached, and suggestions
increasingly explore the implications of At the same time, our understanding of for resources that can complement them.
this technology, from family favourites how we personally interact with AI in our
like WALL-E to the distinctly darker Black day-to-day lives, and how we can use it to AI in action
Mirror. Robotics companies, meanwhile, our advantage, remains limited. Ask a class With the recent proliferation of smart
are beginning to transform the more of 13-year-olds what they think of when speakers and virtual assistants, this
benign of these visions into reality: Hanson they hear ‘artificial intelligence’, for example, technology can be a useful framework for
and the answers tend towards a common an initial discussion around the key tenets
theme: ‘creepy’; ‘sinister’; ‘taking over the of artificial intelligence. Most young people
EMILY world’. How do we retain interest levels will have been exposed to these devices
DREIMANN while grounding AI in reality and preparing in some form; fewer, however, are likely
Emily works for students for the workplace of the future? to identify them as an example of AI. You
the David and Jane For young people to be able to lead and could ask students:
Richards Family succeed in the data-driven economy, a n What does AI look like?
Foundation as project strong understanding of this ever-evolving n What does it sound like?
manager for the technology is paramount. In order to engage n To what extent should it mirror
Get Creative With students with this topic, lessons should human behaviour?
Data course (@ not only highlight the many forms that Google’s Duplex AI assistant is a great
em_dreimann). artificial intelligence can take in the real example of the capabilities and potential
world, but also offer tangible experience of of this technology. In a popular video of the

26 helloworld.cc
 There is a wealth of resources that we can
draw on to assist students in forming their
own opinions in this debate. The Turing GET CREATIVE
test, for example, is an important concept for
students to understand and remains a useful WITH DATA
benchmark against which to measure the
AI and machine learning form a key part of
capabilities of AI technology. Encourage your
Get Creative With Data, a KS3 (ages 11–14)
students to read aloud some of the transcripts
data science course from the David and Jane
for entries to the most recent Loebner Prize,
n Students from Astrea Academy Sheffield saw the Richards Family Foundation (@DJRichardsFF).
production of custom-made sensor boards an annual Turing test competition. Would
any of these convince them that they were Complete course materials are available for
assistant in action, it is heard making calls to speaking to a real human? free to all state schools. For more information
a number of different businesses, sounding On the website AI Weirdness, meanwhile, about introducing the course at your school,
sufficiently humanlike in its interactions to research scientist Janelle Shane publishes contact [email protected].
fool the real humans on the other end of the the entertaining results of her experiments
line. The somewhat unnerving potential of the
AI is likely to hook students, while it remains
grounded in reality as an aid and time saver,
rather than a replacement, for humans. “ AI ARGUABLY REMAINS A THEORETICAL
CONCEPT, WHILE MACHINE LEARNING CAN
Quick, Draw! is a great resource for
highlighting to students another manifestation ALREADY BE SEEN IN ACTION IN INDUSTRY
of AI. The game challenges the user to create
a series of doodles, while a neural network
attempts to guess what they are drawing. training neural networks on existing content is an ideal catalyst for a discussion about
Coupled with its hands-on, accessible nature, across a range of topics, from cat names the differences between how people and
this activity has the potential to engage even to knitting patterns. Taking Halloween machines make sense of instructions. The
the most reluctant of students. costumes as an example, you could allow ability of humans to apply common sense
students to explore; with suggestions such when determining a course of action can be
Not so intelligent? as ‘sentient stone’ and ‘a skunk in a moose highlighted as a strength that machines are
At the same time, we need students to be suit’, it should quickly become apparent to unable to emulate, which can encourage
critical in their appraisal of AI technology. them that this technology has some way to students to understand AI as a tool to
The mantra that ‘Machine learning is written go in capturing the uniquely human traits of complement us rather than compete with us.
in Python; artificial intelligence is written in creativity and humour. Meanwhile, developments in self-driving
PowerPoint’ is a great starting point for this vehicle technology presents a unique
discussion. You could ask students what Branching out opportunity for students to explore ethics
they understand by this. The aim here is to AI and machine learning can also be ideal in the context of computer science. Moral
draw out the idea that machine learning starting points for generating lively debate Machine, developed by the Massachusetts
can already be seen in action in industries around other key topics in Computing. An Institute of Technology, is an interactive
across the globe, while AI arguably remains unplugged activity in which students create tool that asks the user to judge the most
a theoretical concept. Has anyone truly their own algorithm to guide a visitor from acceptable outcomes of a series of moral
created an intelligent machine? the school reception to their classroom dilemmas faced by a driverless car. By
engaging with this modern take on the
classic trolley problem, students develop a
deeper and more personal understanding
of the ethical challenges surrounding
artificial intelligence.
These activities have been popular in our
partner schools, with teachers commenting
that pupils were “fully engaged” and found
the topic “really interesting”. By providing
students with an interactive forum in which
to discuss and explore AI, we have an
excellent opportunity to support the next
generation in confidently claiming their
n Students explored equipment such as circuit-building robots and laser cutters place in the modern world.

helloworld.cc 27
FEATURE

YOU AND AI: RESOURCES

FROM THE ROYAL SOCIETY
Michael Jones explores how AI and machine learning can be
embedded in Computer Science lessons and the wider curriculum

s artificial intelligence the same as in mind, the Royal Society, supported by Teachers in the UK may want to consider
I machine learning? As a computer DeepMind, published ‘You and AI’, a summary using the full-length series for A level
science teacher, I might be expected to of a 2019 public lecture series that explored Computing and Key Stage 5 courses such
have an answer to that question. However, topics as diverse as the future of work, bias, as Cambridge Technicals, for example Unit
the more I consider it, the less sure I am of and how medical diagnosis could be impacted 17 ‘The Internet of Everything’, or BTEC
my answer. The term ‘AI’ is embedded in by AI and machine learning. Computing (Unit 9 ‘The Impact of Computing’).
our language and consciousness, as is, to The full lecture series can be found here: This could be undertaken in conjunction with
a lesser extent, ‘machine learning’. Yet, like royalsociety.org/topics-policy/projects/ projects focusing on the use of AI hardware
me, most people do not have a clear idea machine-learning/you-and-ai. and software such as the Jetson Nano and
of what the concepts mean. Working with Amazon DeepLens.
AI in my role as a teacher, I have some Getting to grips with AI The response to the report and lecture
understanding of the mechanics of AI. For teachers wanting to develop a wider series has enabled teachers from the Royal
Like many, I have a feeling of unease that understanding of AI and machine learning, Society Schools Network to create a body
machines may be making decisions for me. these lectures are a great first stop. Viewed of classroom resources. These come in two
Looking back to the 2018 Facebook– alongside the ‘You and AI’ report (helloworld. forms: short video clips and lesson resources.
Cambridge Analytica scandal, it is all too cc/youandai), they will give anyone
clear that algorithms can be used to influence interested in the topic a balanced and deep Moral decisions in the classroom
how we think and conduct our lives. With appreciation of the technology involved and The short clips from the lecture series (the
this concern about AI and machine learning the challenges it brings. longest is four and a half minutes) each focus

28 helloworld.cc
 on a particular topic. The beauty of these is The lesson resources consist of lesson change?’ package is an example of where
their wide-ranging use. They are suited to a plans spanning Key Stages 3 and 4. computer science and science converge.
cross-curricular purpose; for example, ‘You’ll These are highly detailed and designed The resource is linked to the AQA, Edexcel,
need your hat and coat’ fits neatly into the to engage and raise interest. The KS3 and OCR science specifications. Linked
wider science curriculum focusing on how AI package ‘How do we stop machines to the ‘Practical applications of AI’, the
can be used to forecast weather. Clips such enslaving us?’ provides students with resource investigates ocean acidification
as ‘What kind of world do we want?’ and an opportunity to investigate decision- and looks at how AI can be harnessed to
‘When is a criminal not a criminal?’ provide making algorithms and the life-and-death provide us with greater understanding and
a PSHE focus. The relatively short length consequences they may have. I encourage perhaps with the means to prevent the
of the clips and their ‘big question’ focus you to test your moral judgement using damage occurring. The resource is supplied
make them well-suited to debate and starter the MIT Moral Machine (moralmachine. with ready-to-go examination questions.
scenarios. As with most ‘big question’ topics,
there is likely to be a diversity of answers.
That can only be to the good and helps to
ensure that our students are exposed to
“ MOST PEOPLE DO NOT HAVE A CLEAR IDEA
OF WHAT AI AND MACHINE LEARNING MEAN,
questions and ideas that will undoubtedly
impact their lives. DESPITE THEIR PREVALENCE ISOCIETY

mit.edu). Starting from an individual Supporting the videos and lesson

YOU AND AI viewpoint, students contribute to a wider packages are an informative animation

RESOURCES group rubric about the value of lives and

the meaning of responsibility.
and two interactive infographics. Give the
identification activity a go and see if you can
The uses of neural networks and beat Google. The immediately accessible
Lecture series summary: helloworld.cc/youandai
classification are explored in the ‘How do content in the infographics will work well
Video clips: helloworld.cc/society
machines learn?’ package. Here, students from Key Stage 2 upwards. They act as
Teaching suggestions: helloworld.cc/resources
learn what an algorithm is and how stand-alone materials in which students can
Interactive infographic: helloworld.cc/info classification rules are weighted. The explore the topics and questions and test
example given focuses on the classification themselves as they progress.
of butterflies (see p.83 The Royal Society is committed to
for more on this) and the supporting schools, not just in computer

You
features that may lead to an science but as part of a much wider science
object being classified as a remit. The Schools Engagement team has

and AI
butterfly or not. The resources a growing schools network across the UK,
are suited to adaptation, providing opportunities such as access to
recognising facial expressions funding opportunities for student research
Conversations about AI technologies
and their implications for society or trees versus teacups, for projects, student conferences, and lesson
example. Mathematical skills resources at both primary and secondary
are incorporated in this via level. To become part of the network
the concept of weighting a and receive their newsletter, sign up at
decision: if a feature exists, it royalsociety.org/schools-network.
assumes a weight of 1. Non-
existence assumes a weight
of 0. The closer the average MICHAEL
of the weights is to 1, the JONES
more likely it is that the object Michael is Director of
is a butterfly. This is a highly Computer Science at
practical activity and could Northfleet Technology
be used as an unplugged College. He is a
classroom activity. Specialist Leader of Education, a Raspberry
Pi Certified Educator, and CS Champion
Taking Computer Science for the NCCE (@MikeJonesCSTalk,
into Science at GCSE community.computingatschool.org.
uk/users/222).
SUPPORTED BY
At Key Stage 4, the ‘Can AI
CONVERSATIONS ABOUT AI TECHNOLOGIES AND THEIR IMPLICATIONS FOR SOCIETY
DeepMind 1

predict the future of climate

helloworld.cc 29
RESEARCH

#INSIGHTS
STORYTELLING IN
COMPUTING EDUCATION
STORY BY Hayley Leonard

hildren’s traditional stories, nursery is happening and introduces computing The program step is supported by the
C rhymes, and literature provide a terminology. Children then act out the story sample lesson resources developed for
rich source of sequences and repetition. and are asked about repeating patterns specific books. It involves using the
At the 14th Workshop in Primary of behaviour and what triggers them. computer to produce the program that
and Secondary Computing Education The model element of the approach then has been designed through the previous
(WiPSCE 19) in Glasgow last October, involves constructing the sequence of unplugged activities.
Sarah Twigg and colleagues presented an events in the story, using either images
approach that uses children’s literature to from the book or printouts of lines of code Teachers’ evaluation of the resources
teach Computing to primary school pupils. or blocks from Scratch. Children are asked Responses from teachers who were
The team from Lancaster University to identify repeating patterns and choice asked about their experiences were
hopes the familiar contexts of children’s points in the story, and links are made to highly positive. In particular, teachers said
stories will engage pupils and raise the the computing terminology throughout. that the first three stages (read, act, and
confidence of non-specialist teachers in
delivering the curriculum.

Identifying computing concepts LESSON EXAMPLE

in children’s literature
The team reviewed 50 popular children’s Computing concepts represented in We’re Going on a Bear Hunt (Rosen, Oxenbury 1989):
picture books to identify key computing
Sequencing: A list of events to be followed order. On their way back home, they go
constructs, namely sequencing, repetition, in order. through the environments in reverse order.
and selection. The books included
Example: The characters on the bear hunt Repetition: At least one example of a
programming constructs to varying go through six different environments in pattern of repeated dialogue, actions,
degrees, but 16 books included all three or environment.
constructs, and all 50 books could be used Example: Repetition of particular phrases
to demonstrate sequencing. Several books in each environment, for example “We’re
were then used as the basis for sample going on a bear hunt.”
teaching resources, which were trialled in
some classrooms and code clubs. Selection: At least one example of a choice
of dialogue, actions, or environment.
The teaching approach: Read, Act, Example: The end of the repeated dialogue
Model, and Program (RAMP) varies depending on the environment,
for example, in the river they say,
An approach named RAMP builds up “Splash splosh!”.
subject knowledge and appropriate n Text © 1989 Michael Rosen, Illustrations © 1989 Helen Oxenbury,
vocabulary in a storytelling context. The From WE’RE GOING ON A BEAR HUNT by Michael Rosen. Reproduced
by permission of Walker Books Ltd, London SE11 5HJ,
format begins by reading through the story; www.walker.co.uk

the teacher asks questions about what

30 helloworld.cc
 “ THE APPROACH
IS LOW-COST
AND USES
FAMILIAR
CONTEXTS
model) were very engaging for the pupils.
They provided multiple opportunities for
differentiation and working together at
different levels of ability.
Some teachers suggested, however,
that some non-specialist computing
teachers might find the step up to the
program stage daunting. The authors
are using this feedback to help them
develop the teaching resources. They
are continuing to work with teachers
to investigate whether more support is
required for the transition to the final
stage, or whether this should be separated
from the other elements of the approach.

An inclusive approach
Teaching computing principles to young
children can be a challenge. A creative
storytelling approach is low-cost and uses
familiar contexts that are intuitive to teachers
and parents. Twigg and her colleagues
suggest that this approach has benefits for
diversity in computing: they are particularly
interested in its use with disabled students.
The collaborative activities and active
discussion involved could also be beneficial
in engaging more girls in computing. A pilot
FURTHER READING
study that tests this approach more widely
will start in England in September 2020, as Twigg, S., Blair, L., and Winter, E (2019), Using children’s literature to introduce computing
part of the Gender Balance in Computing principles and concepts in primary schools: work in progress (helloworld.cc/literature)
programme of research.

helloworld.cc 31
RESEARCH

VIRTUAL TEACHING ASSISTANTS

HELP STUDENTS LEARN
STORY BY Thom Kunkeler

earning in classrooms is usually instruction. In this respect, open learner four groups, with students who had access to
L guided by a teacher, but new models are learning environments that allow the open learner model and control over their
research has shown the promise of using students to access content with varying problem selection, and students who did not.
virtual assistants. A study has tested levels of interactivity. For this study, Long and The researchers found that students who
the effects of an open learner model, in Aleven investigated how an open learner learned with the open learner model needed
which students can view and reflect upon model can support students’ self-assessment significantly fewer steps and problems
their level of mastery in different aspects of equation-solving skills, and whether shared to reach the mastery criterion, and had
of problem-solving. Students can then control over problem selection ensures better fewer incorrect attempts. In addition, the
choose which problems to work on in learning gains and enjoyment of learning. self-assessment prompts might have led
order to progress. Yanjin Long and Vincent They designed a user interface using students to reflect on the skills they just
Aleven, who conducted the research in Cognitive Tutor Authoring Tools (CTAT) in practiced, or stay more focused on practising.
two US classrooms, found that students which each problem is broken down by the The presence of an open learner model can
who had control over their learning in this student into steps, providing step-by-step therefore lead to significantly better learning
way showed significantly better equation- guidance. The tool was also used to provide gains when shared control over problem
solving skills and enjoyment levels than feedback for each step, and students could selection is granted to the student.
their counterparts. request hints for what to do next.
Over the past decades, the field of As learners advanced through the
Intelligent Tutoring Systems has emerged
with the goal of investigating how artificial
problems, their skill bar progressed, turning
gold when students completed over 95% of “ STUDENTS
ACCESS
intelligence techniques can be beneficial to the assignments. In addition, the researchers
the personalised needs of individual learners, added self-assessment prompts where CONTENT
while helping teachers out in classroom students could express their confidence in
solving a type of problem. To understand how
WITH VARYING
the open learner model relates to learning INTERACTIVITY
VIRTUAL TEACHING outcomes, the classrooms were divided into

ASSISTANTS
n Cognitive Tutor
Authoring Tools:
helloworld.cc/tutor

n Lynnette:
helloworld.cc/smarter

n MagicBox: helloworld.cc/box

FURTHER READING
Long, Y. and Aleven, V.
(2017). Enhancing learning
outcomes through self-
regulated learning support
with an open learner model n Lynnette is an intelligent tutoring system
(helloworld.cc/open) with an open learner model that displays
skill mastery in the forms of skill bars

32 helloworld.cc
 ACADEMIC INSIGHTS FROM THE
WIPSCE CONFERENCE
STORY BY Oliver Quinlan

ne of the key events for Computing The conference started with a focus programming through the lens of another
O Education research in Europe is on primary education, and several papers subject, and discussing this with others,
the Workshop in Primary and Secondary explored young children’s perceptions was a good chance to be put in the position
Computing Education (WiPSCE) which about programming and technology. There of the students we teach.
took place last October in Glasgow. Sue is a lot of rhetoric about children growing As with any conference, it isn’t just the
Sentance and I, from the Raspberry Pi up around technology and developing formal sessions that are valuable, but also
Foundation, headed to Glasgow to hear expertise with ease, but the authors of the chance to meet and discuss with new
from colleagues in computing education these papers took an objective view of how people working in similar, but different areas
and share some of our work. children’s perceptions are developing with to your own. WiPSCE was a very friendly
Academic conferences have a particular some very interesting discussion. conference in this sense, and I enjoyed
format and although they often have On the second day I presented the work some thought-provoking conversations from
keynote speakers covering big issues, they we have been doing at the Raspberry people working on school-based computing
largely involve participants presenting Pi Foundation, exploring approaches to education across the world. Despite it being
their own work and inviting feedback and assessment during informal learning. an academic conference, there were also lots
discussion on it. Presenters submit papers Presentations at academic conferences of practising teachers at the event, which
about their research projects and findings, are often met with more challenge than brought a real intersection of views from
and then present the content of these at other types of events, with much theory and practice together.
papers for other presenters and attendees of the value coming from discussing For educators, academic conferences
to learn from but also to critique. and developing the ideas. The diverse might seem like something rather different,
WiPSCE’s two keynotes started each perspectives of attendees were really but those like WiPSCE can provide
day off with a big issue. On day one, Beth helpful in pointing out some aspects of our experiences to engage with the deeper and
Simon from the University of California paper that I hadn’t considered, and I found more emerging themes in your subject. You
raised the big questions about training a this constructive challenge very helpful. will rarely find quick fixes or off-the-shelf
new generation of computing teachers, As well as paper presentations there resources at such conferences, although
and how we should be looking to the were also workshops, and I attended the there were some tools on offer here which
future. On day two, my colleague, Sue session focusing on ScratchMaths. This I could see being used in the classroom by
Sentance, identified the potential for project aims to teach maths skills through enquiring teachers. What you will find are
embedding computing into a national computing, and specifically programming in presentations that really make you think
education system, with the case study Scratch. Celia Hoyes showed us how they about the underlying issues that our subject
of the National Centre for Computing have developed materials to successfully faces. You can find out more about future
Education in England. address both subjects. Looking at WiPSCE conferences at wipsce.org.

n WiPSCE 2019 was held at Glasgow University

helloworld.cc 33
 SUBSCRIBE
SIGN UP TODAY FOR FREE
=

d
for UK-base
educators

SUBSCRIBE
TODAY
 ublished four
P
times a year
 ave them
H
delivered directly
to your door
 ello World is not
H
available in stores

34 helloworld.cc
 TO SUBSCRIBE VISIT:
helloworld.cc/subscribe
Not a UK-based educator?
Buy in print for £6
Visit: store.rpipress.cc

FREE
P DF
for anyon
e,
anywhere

helloworld.cc 35
 FEATURE

COMPUTING AND THE REVISED

OFSTED INSPECTION FRAMEWORK
Ofsted’s new inspection framework emphasises ‘quality of education’.
Neil Rickus examine what this means for teachers, subject leaders, and pupils

chools in England have been outlines how schools should focus on the and success may not be purely academic.
S inspected under a revised Ofsted intent, implementation, and impact of the The new framework also includes an
inspection framework since September curriculum, which can be briefly defined as: emphasis on cultural capital, with the
2019 and, while there have been various inspection handbook stating:
reports about what an inspection n Intent: The focus here is on the “It is the essential knowledge that pupils
now entails, including subject-specific curriculum and aspirations for learners. need to be educated citizens, introducing
‘deep dives’, only a few schools have All learners should study an ambitious them to the best that has been thought and
actually been inspected under the new curriculum that is rich in both knowledge said and helping to engender an appreciation
framework. In this article, we discuss the and skills and offers suitable progression of human creativity and achievement.”
key changes to the framework and share across the school. Within computing, this could perhaps
the details of a visit to Wembley Primary n Implementation: This is about how mean focusing on:
School in London to find out about their teachers do their job and how leaders
recent experience of being inspected. support them, and focuses on assessment n Technology: How it can be disruptive and

and feedback, pedagogy, subject improve lives, along with negative impacts
Quality of education knowledge, and the quality of teaching. n People: Ada Lovelace, Alan Turing,

Chief Inspector Amanda Spielman recently n Impact: This looks at how learners Grace Hopper, Bill Gates, Steve Jobs, etc.
outlined the need for schools to have a develop their knowledge and skills across n Opportunities: Extracurricular activities,

broad and balanced curriculum, which is the curriculum and, as a result, their going on school trips, entering events
reflected in the new judgement category achievement. It is less focused on data and competitions, talks by or visits to
of quality of education. This category than the previous inspection framework, employers and educational providers

36 helloworld.cc
Intent n Eleanor Haines and Satya Maremanda are computing
coordinators at Wembley Primary School, London, which
In order for schools to have a broad and had a recent Ofsted inspection under the new framework
balanced curriculum, foundation subjects
such as computing need to be studied
at appropriate depth and given sufficient
curriculum time. In order to discuss and
evidence how computing is delivered,
thorough and accessible schemes of work,
including the progression in skills and
knowledge, could be part of a subject leader
folder (see box on p.38). Cross-curricular
links and opportunities to develop cultural
capital could also be identified.
To find out more about the realities
of a recent inspection, I visited Eleanor
Haines and Satya Maremanda, computing
coordinators at Wembley Primary School
SUBJECTS SUCH AS COMPUTING NEED TO
(WPS) in London, whose September
2019 Ofsted report (helloworld.cc/ “ BE STUDIED AT APPROPRIATE DEPTH AND
ofsted) outlined how well-thought-out
the school’s computing curriculum is GIVEN SUFFICIENT CURRICULUM TIME
(see p.2).
At WPS, a curriculum map (see photo on
p.38) specifies the topics and technologies Time is spent in other curriculum are regularly sought, with online safety
used by each year, per half term. As some subjects revisiting various key skills, being discussed with parents at a range
staff members lack confidence in delivering although coordinators work closely with of events taking place throughout the
computing, lesson plans with step-by-step staff to ensure that technology is being academic year.
instructions are also provided, which link used appropriately, rather than just for the To enrich the curriculum, WPS holds
directly to the programming environment sake of using it. Online safety content is computing clubs for both Key Stage 1
used throughout the school. Providing predominantly delivered within computing and Key Stage 2 pupils, which are well
planning in this way helps teachers focus lessons at WPS, with a focus at the start attended and provide an opportunity for
on the required skills and knowledge of every academic year on the school’s all pupils to use the school’s technology
within each session, rather than the acceptable use policy (AUP). The views of away from the formal classroom
underlying technology, to avoid ‘We’re both pupils and parents on the computing environment, along with occasional
doing Scratch today.’ curriculum and technology usage in school enrichment days.

NEIL RICKUS
Neil is a Senior Lecturer in Computing
Education at the University of Hertfordshire,
a primary education consultant for the BCS,
and the founder of Computing Champions.
He is a CAS community leader, a CEOP
Ambassador, and a Raspberry Pi, Google
and Microsoft Certified Educator
n Knowledge organisers provide information
for pupils on the topic being studied, such as (@computingchamps).
relevant vocabulary and suitable code blocks

helloworld.cc 37
 SUBJECT LEADER
FOLDER
A detailed, well-organised subject leader folder
can enable intent, implementation, and impact
to be evidenced, along with outlining how the
subject is to be developed and facilitating
discussion during an inspection. Many CAS
community meetings this term will focus on
the new inspection framework and enable
attendees to discuss subject leadership in
their school.
The contents of a subject leader folder
could include:
n Development/action plan
n Wembley Primary School’s computing curriculum
n Observations of lessons map outlines the topics studied and the
technologies used within each year group
n Self-evaluation
n Professional development records
ALL LEARNERS SHOULD STUDY AN AMBITIOUS
n
n
Staff audits and surveys
Pupil and parent questionnaires “ CURRICULUM THAT IS RICH IN KNOWLEDGE
Budget and resource documents
AND SKILLS AND OFFERS PROGRESSION
n

n Progression documents
n Schemes of work
n Cross-curricular topic links Further guidance is provided by the use Ofsted framework as the focus for their
n Moderation/examples of students’ work of knowledge organisers, particularly next event. Details of your nearest
n Online safety guidance for programming units of work, which event can be found here: community.
outline the relevant blocks of code and computingatschool.org.uk/events.
vocabulary examined within lessons (see CAS have created a Next Steps document,
Implementation photo), along with scaffolding within the which includes a number of tasks for subject
The implementation of the curriculum programming environment itself, such as leaders and possible questions they may be
appears to be focused on what happens tutorial videos. asked during an inspection. The document
in classrooms on a day-to-day basis, can be found within the CAS resources
rather than examining paperwork or Impact pages: community.computingatschool.org.
consulting with the Senior Leadership Finally, the impact of the curriculum uk/resources/.
Team (SLT). For example, inspectors examines pupils’ achievements in school, A more extensive interview with
may wish to view computing lessons which is related to how they develop their Eleanor and Satya from Wembley Primary
delivered by a non-specialist teacher, skills and knowledge across the curriculum. School is available on the CAS website:
or to discuss the subject with pupils. Schools could consider how they can computingatschool.org.uk/.
The coordinators at WPS outlined how evidence pupils’ work and link this to the Has your school recently been inspected
inspectors were keen to find out more relevant skills and knowledge. End points under the new framework? Perhaps the
about the computing lessons taking place, for pupils could also be identified, such inspection outlined above differs from your
including discussing with teachers where as programs and content produced by experiences? Maybe you lead computing in
the current lesson fitted into a scheme of children of varying abilities. For example, a different way and would like to share
work, along with previous content taught the inspection report for WPS praised the your approach? Do get in touch on Twitter
and the next steps in pupils’ learning. school for enabling Year 6 pupils to use their at @computingchamps.
To enhance teachers’ subject knowledge of computing to design their
knowledge, regular continuing professional own games (see p.2). Disclaimer: The information outlined in this
development is provided at WPS, along article does not reflect the official views of
with support from the coordinators, other Next steps Ofsted, Computing at School (CAS), or the
staff and digital leaders (a small group of Many Computing at School (CAS) authors, is subject to change and should be
pupils confident in computing) as required. community meetings have the revised used for guidance only.

38 helloworld.cc
 HELPING BUILD THE
ENGINEERS OF TOMORROW
Stew Edmondson explains how Electronics
Everywhere is helping students develop the computing
skills required for future careers

he UK has a great heritage of

T technological innovation and a
world-class electronics sector. However,
there is now a shortage of electronics
engineers, meaning that there are too
few engineers and designers to develop
the next generation of products and help
produce creative technological solutions
needed by society. The UK Electronics
Skills Foundation (UKESF) wants to
encourage more young people to consider
engineering as a career, and to be aware
of the opportunities available in the
electronics sector, for the benefit of the UK
and wider society.
We know that computing is an
increasingly popular A-level subject with
n Students working with the Logic and Arithmetic Kit
10,000 students nationwide sitting the Credit: University of Southampton

exam in 2018; this was up 23 per cent on

2017. Despite this, there are few hands-on increase their knowledge about electronics
teaching resources available to schools to and so inspire more young people to take
teach fundamental computer engineering. up related careers.
Therefore, working closely with academics Electronics Everywhere has received
from the University of Southampton’s formal endorsement and support from the
prestigious School of Electronics and Royal Academy of Engineering (RAEng).
Computer Science, the UKESF has The UKESF is now working with RAEng to
developed a Logic and Arithmetic Kit. The promote the project via their Connecting
kit features hands-on activities with a circuit STEM Teachers programme and using
board that teach core electronics concepts their teacher coordinators to improve the
to A-level Computer Science students. learning resources.
Along with the Music Mixer Kit, aimed
at A-level Physics students, the Logic and Resources for teachers
Arithmetic Kit is part of the Electronics The Electronics Everywhere boards have STEW EDMONDSON
Everywhere project, which aims to show been professionally manufactured and Stew has been CEO of the UK Electronics
young people how engaging electronics supporting resources, including video Skills Foundation since 2015. He is a
can be and the exciting career opportunities tutorials, have been produced. Kits have Chartered Engineer and a Fellow of the
that are available within the sector. already been distributed to over 170 Institute of Engineering and Technology. He
Importantly, the project also provides schools and one teacher from each school has a degree in Electronics (@theUKESF).
professional development for teachers to has received face-to-face training at the

helloworld.cc 39
 FEATURE

NOT), and three logic outputs. Orange

LEDs indicate the state (logic 0 or 1) of the
UNIVERSITY OF SOUTHAMPTON SUPPORTS individual inputs and outputs of every gate.
The Arithmetic section provides an
ELECTRONICS EVERYWHERE 8-bit two’s complement circuit for adding/
subtracting, offering the ability to perform
A+B or A-B (where A and B are 8-bit binary
Encouraging A-level students in computing and
numbers). This can be used to experiment
electronics can be a challenge. When young
with unsigned and signed binary arithmetic,
people visit our university on open days, we really
as well as offering a different way to
need to pull it out of the bag to ensure they have
observe and understand binary number
the best possible experience. Programmes of
systems. As an extension, a 9-bit adder
events can consist of a sample lecture, a practical
circuit can be created, by combining both
session, a tour of the campus and a curriculum
section of the circuit board.
targeted activity. As an electronics and computer
The kits have been used by over 700
science department we were concerned about
the lack of electronics and computing outreach students so far; 70 per cent reacted ‘very
and resources aimed at A-level pupils. This people to consider engineering as a career and positively’ and were more enthusiastic
is especially as there is now a shortage of to be aware of the opportunities available in the about electronics as a result. Of the teachers
electronics engineers (especially graduates), electronics sector for the benefit of the UK and using the resources, 57 per cent said they
meaning that there are too few engineers and wider society. were an ‘excellent’ teaching aid. Comments
designers to develop the next generation of One fantastic aspect of the project is that it included, ‘The kits made a real difference
products and help produce creative technological provides outreach in the form of professional to the engagement of the students and a
solutions needed by society. development to teachers. It empowers them with quite tricky topic was made much more
The UK Electronics Skills Foundation (UKESF) the resources to teach aspects of A-level Computing accessible’; ‘Students have a much better
and the Electronics and Computer Science they may have not enjoyed doing previously. appreciation of data representation, memory,
Outreach Department at the University of storage and decision making as a result of
Southampton want to encourage more young - Reena Pau, University of Southampton using these’; with further comments noting

ELECTRONICS
“ EVERYWHERE
SUPPORTS
TEACHERS AND
SCHOOLS

that ‘They are great bits of kit that bring alive

the concepts of logic’.

How to get involved

Electronics Everywhere is a high-impact
n The Logic and Arithmetic Kit circuit board
Credit: University of Southampton project. We are investing in and supporting
teachers and schools at a local level
University of Southampton. 2 (binary) number systems. These topics right across the UK. The project provides
The Logic and Arithmetic Kit’s bare are most often taught through simulation reusable classroom resources and trains
printed circuit board exposes all the or PowerPoint. The board is split into two teachers in their use. This allows teachers to
components, creating a link between sections: Logic and Arithmetic. deliver the A-level curriculum for computing
electronics and applications in many The Logic section can be used to explore in an engaging and interesting way and so
consumer products. This allows students Boolean operations, truth tables, and improves the learning experience for their
to pursue their curiosity, showing that logic gates, and to implement simple logic pupils. Each school involved in the project:
electronics is not a mysterious black box. functions and circuits. The board has three n Receives a classroom set of circuit board

The kit is reusable and covers aspects of switchable logic inputs, a range of different kits; these resources are fully packaged
Boolean operations, logic gates, and base logic gates (AND, OR, NAND, NOR, XOR, and are reusable

40 helloworld.cc
n A Computer Science teacher running a session with the
Logic and Arithmetic KitCredit: University of Southampton.

n Has a teacher complete a comprehensive The ultimate aim of Electronics Electronics Everywhere is also linked
training session to familiarise themselves Everywhere is for every school and to a broader campaign from the UKESF
with all aspects of the teaching resource college in the UK which offers A-level called #TurnOnToElectronics. This
(a video tutorial is available for teachers Physics and/or Computer Science to have multiplatform campaign is fronted by
unable to travel to Southampton) a classroom set of the kit, supplied free The Gadget Show presenter Georgie
n Gains access to online teaching of charge to those in the state sector. If Barrat and aims to encourage more
resources, lesson plans, guides, and you would like to get involved, please young people to study and pursue
additional information visit ecs.soton.ac.uk/outreach/kits where careers in electronics. Georgie said, “I’m
n Is covered by a comprehensive support you can sign up to the mailing list or delighted to be involved in the UKESF’s
package, including a replacement/repair email Southampton directly with specific #TurnOnToElectronics campaign, which
service. queries about the project. is shining a welcome spotlight on careers
in Electronic Engineering… studying
electronics can lead to an exciting and
creative career, which enables students to
UK ELECTRONICS SKILLS FOUNDATION make a real difference in the world.”
To find out more, visit
The UKESF’s mission is to encourage more young turnontoelectronics.org, where you can
people to study electronics and to pursue careers in watch the campaign video, download the
the sector. The UK electronics sector is big, valuable, manifesto, and access a wealth of
and growing; however, the demand for capable, information targeted to young people —
employable graduates is currently outstripping supply. from what it is like to be an electronics
The UKESF is a charity and operates collaboratively engineer in the UK to where to go to
with major companies, leading universities, and other develop an interest in electronics. Follow
organisations to tackle the skills shortage in the sector. @TOTElectronics to keep up to date with
the campaign.

helloworld.cc 41
 FEATURE

CRACKING IDEAS: A COMPETITION

FOR YOUNG INVENTORS
Bruce Robinson shares how the Intellectual Property Office and Aardman
Productions have collaborated to encourage creative ingenuity

something that protecting their new intellectual property:

will make life this includes patents, trademarks, and
better. There copyright. Children learn the value of their
are resources to original ideas and the inventions they create,
support teachers how to protect them from exploitation, and
in delivering this how to eventually convert their creations
competition with into a commercial product.
their students, The demand for inspired ideas and
as well as inventions isn’t going to change anytime
activities for soon. The Cracking Ideas competition
students up to gives children the opportunity to exercise
the age of 18 their creative muscles and forge the future
and beyond. devices they are going to live with one
A Cracking day, and shows how it is possible to make
Ideas project money out of their ideas, too.
can be big or The competition is free to enter, and the
small and can resources are available for free after
olutions to the world’s trickiest encompass any area of creative STEM; it is registering at crackingideas.com. The
S problems sometimes result from up to the students and their teachers what winner’s invention will star in a brand new
years of careful research and product the project is and where it is going. It could Wallace and Gromit digital project in
development, sometimes they rise out of be used to develop a Big Bang Exhibit or an 2020. They will also win an Aardman
looking at a failed project in a new way, entry for a Crest Award, or be part of the model-making workshop for their school
and occasionally they just evolve from a Children’s University Passport scheme. or group. The competition closes on Friday
germ of an idea following a discussion of The resources encourage students to learn 24 April 2020.
a far-reaching idea that’s been bounced and research as they go, discovering as they
around between a team. advance their ideas. They learn to appreciate
Young people see the world around them real-world limitations and push boundaries
without boundaries. This creativity can be a with their imaginations.
powerful tool for invention, because young Cracking Ideas materials are designed to
minds are often willing to push the limits, help learners develop the design process,
simply because they don’t recognise the iteratively improving through prototyping,
limits; they haven’t learnt them yet. and then produce their invention, whether as
a clay model, an engineering illustration, or a
Creativity and intellectual property working piece of electronic wizardry. Nothing BRUCE ROBINSON
The Intellectual Property Office (IPO) and is out of bounds and nothing is wrong, Bruce is a STEM presenter with Wonder
Aardman – the animation company behind because what is produced comes straight Workshops. He is a retired teacher and a
Wallace and Gromit – have launched a new out of the minds of our young inventors. professional development provider with
competition that encourages children aged In the final stage of the process, the STEM Learning and Science Learning
five to eleven to harness their creative ideas. children take ownership of their idea or Partnership (@wonderworkshops2).
To enter, young inventors need to design invention by learning about the process of

42 helloworld.cc
 n Many Schoolhouses have seen an increase in girls opting to take
computing-related subjects since joining the programme

DIGITAL SCHOOLHOUSE:
COMPUTING THROUGH PLAY
Estelle Ashman explains how hosting a programme of workshops and CPD to inspire
local primary students in creative tech boosted her own confidence and knowledge

s Digital Schoolhouse’s Curriculum here I was with like-minded individuals from With a catalogue of workshops
A Content Developer, Estelle Ashman all over the country being trained by some of that include everything from lessons
is responsible for the creation and curation the best people in computing pedagogy and on algorithms using Ubisoft’s Just
of their innovative teaching resources. I was being told that no idea was too crazy! Dance to investigating binary numbers
With over ten years’ teaching experience Not only that, but I was being paid to be there with Nintendo’s Super Mario Maker
and a master’s degree in Teaching and – Schoolhouses can apply for a bursary and 2, Digital Schoolhouse is at the
Learning, Estelle is an enthusiastic travel for all training is reimbursed. forefront of playful computing.
practitioner who has the expertise to take
difficult computer science concepts and
present them in a new and innovative way.
Estelle’s role with Digital Schoolhouse MACHINE CODE MARIO
is hybrid. Her working week is shared
between Digital Schoolhouse and Digital Schoolhouse is known for its
classroom teaching at Gildredge House free, innovative, playful workshops.
in Eastbourne. In this example, students are taught
It was almost five years ago when I how to create Super Mario levels that
submitted an application for my school to require the player to carry out binary
join Digital Schoolhouse and what a brilliant conversions and calculations in order to
five years it has been. At the time, I wasn’t complete the level.
completely sure what I was joining. Download the workshop resources
My first experience of a Digital Schoolhouse here: helloworld.cc/mario.
training event was an innovation – finally,

helloworld.cc 43
 FEATURE

HOW DIGITAL SCHOOLHOUSE WORKS

The not-for-profit Digital Schoolhouse programme, n The aim of each workshop is to teach pupils
together with Nintendo UK, uses playful learning computing in a way that is creative, innovative,
to engage the next generation of pupils and inspirational, engaging, and fun.
teachers with the computing curriculum. Digital n Computer science concepts are taught using
Schoolhouse is delivered by UK games industry play-based learning. They fuse art and science
trade body Ukie and supported by the Department through dance, games, magic, and storytelling.
of Digital, Culture, Media, and Sport (DCMS).
n Each digital schoolhouse lead teacher is
The process works like this: given bespoke training to improve their subject n Digital Schoolhouse prides itself on providing high-quality training
n Education settings (typically secondary knowledge and teaching pedagogy. led by some of the top practitioners in the country

schools and FE colleges) are selected n All visiting teachers are provided with free
to become Digital Schoolhouses. and personalised support following the visit. of local primary and secondary teachers
n They offer free weekly workshops to visiting This helps them continue to embed computing to deliver creative and cross-curricular
primary school pupils and teachers. in their schools. computing lessons using playful learning.
Through this model, the programme
supports the computing programme
of study for the national curriculum in
a way that leaves pupils and teachers
feeling inspired about, and engaged with,
computing and the wider creative digital
industries. By joining the programme as
a Digital Schoolhouse, it’s also possible
to improve the transition from primary to
secondary school, and secondary teachers
can have a better idea of the new intake
while raising the profile of computing within
their schools and local communities.
Applications are now open for schools
wanting to join the programme. For more
information visit digitalschoolhouse.org.
n By joining the programme, Schoolhouses have access to a
network of teaching resources, including Nintendo LABO kits uk/join-dsh.

Rewarding relationships impact on my sense of worth — it is

Working with local primary schools is so hugely rewarding to know that someone
rewarding, from encouraging students is noticing your hard work — but it
to work in groups, which I am often told also opened our school to even more
doesn’t happen as often as it should, opportunities. Over the last few years our
to engaging disengaged students, students have had the chance to run their
and every session is different. Each own national e-sports tournament, visit
workshop has one thing in common: top game studios, take part in competitions
innovative resources that bring the and I have had the chance to talk about
curriculum to life through playful learning. the programme at events and work with
With our involvement stretching over partners at trade shows. In the last year,
five years we are now seeing the I have had the further opportunity of
positive impact that the workshops becoming part of the team and now split
have on both the students that go on my time between working for Digital ESTELLE ASHMAN
to study with us and the development Schoolhouse and teaching at school. Estelle Ashman is a teacher at Gildredge
of our own teaching resources. House, Eastbourne, and Digital Schoolhouse’s
I have Lea Gilbert, the previous Head of Join the programme Curriculum Content Developer, responsible for
Gildredge House, to thank for encouraging Each Digital Schoolhouse is based in a the creation of their teaching resources
me to join Digital Schoolhouse. Not school, college, or university environment, (@CompSci_Geek).
only did her suggestion have a positive and aims to work with a growing network

44 helloworld.cc
LEARNING LANGUAGES WITH SCRATCH
Scratch and other programming languages are handy educational tools for learning languages

s the Spanish and Computing to apply logic to their actions, which

A coordinator at my school, the helps develop their reasoning skills. They
Premier Academy in Bletchley, I have work in pairs and small groups, which
been trying to use programming activities encourages collaboration, an essential
to improve my students’ Spanish skills. skill for everyone in today ́s society. As
Like other teachers, I have been very the learners need to record words and
impressed with the capabilities of Scratch, sentences to add to their projects, they
a block-based programming language and improve their speaking and listening skills.
online community. With Scratch, children Recording themselves and listening back
can program and share interactive media as many times as they like can help to
such as stories, games, and animations; it identify errors in pronunciation and boost
offers students opportunities for deeper their confidence. Here are a few of the
learning and understanding of their work ways I’ve incorporated programming into
in different areas of the curriculum, such as my Spanish lessons. JESUALDO MARTÍNEZ
Maths, Art, Science, History, and Modern is an Assistant Head Teacher and Computing
Foreign Languages (MFL). Animated illustrations and Spanish Coordinator at the Premier
As MFL teachers, we can use Scratch to Academy in Bletchley, home of the
Why combine programming with MFL? improve children’s vocabulary skills. They codebreakers. He is the author of the blog
By combining programming with can design animated illustrations about the ‘Teaching Languages with Technology’
Spanish, the children are forced to think different topics we have previously worked (helloworld.cc/tech).
creatively and systematically; they have on in our lessons, such as body parts,

helloworld.cc 45
 FEATURE

WITH SCRATCH,
“ CHILDREN CAN
PROGRAM
STORIES, GAMES,
AND ANIMATIONS

classroom objects, or food. Once they have

prepared the images, they add text to the
different objects so that they can practise
their spelling skills; finally, they record their
voices, which also helps them to improve
their pronunciation skills. In this way, my
students have designed cookbooks and
interactive body part illustrations.

Interactive games
Another strategy is to ask the children
to design interactive games that use the
vocabulary they are learning in the lessons,
for example making games in which
they click on the different words they
are listening to, or quiz games in which
they have to answer real questions about
the topics they have been working on.
Something that is a little more difficult, but
extremely rewarding, is to create games
in which the player clicks on words in the
correct order to make a sentence. If you

46 helloworld.cc
 The end result
Incorporating these
programming activities
into my Spanish
lessons has been very
motivating and positive,
not only in developing
digital literacy, but also
in improving students’
language skills. This
has been reinforced
by research. Studies
have shown that the
use of programming
activities can be a good
way of improving both
academic results and
student motivation.
As the Computing and
Spanish Coordinator at
my school, I always try
click the wrong word it turns red, while Without realising, they are learning by to use technology as an integral part of my
it turns green when it is correct. Another themselves by researching information language lessons. Scratch is just one of the
great game is ‘Match Pairs’, which can be in their books and dictionaries. They can many tools that educators can use to make
played with pictures, words, or both and choose any character they like, switch lessons more motivating, collaborative,
involves exactly what the name suggests. backdrops, make their characters have a exciting, and engaging for our students and
This is a great way for children to learn conversation, move the characters, and their different learning styles.
new vocabulary through play. change an element in the story when they Scratch is free, and with a teacher
click on it. It’s guaranteed to be fun! account we can create accounts for
Animations and interactive stories Once their projects are finished, the groups of students and manage their
My students love to code their own students share them with their partners projects. The Scratch website has many
animations and interactive stories, and so that everyone in the class can enjoy resources, tutorials, and ideas to try in
this contributes greatly to improving their and benefit from each other’s work. various areas of the curriculum.
grammar and communication skills. It
promotes creativity, collaboration, and
engagement. This allows them to practise
writing skills as they prepare a storyboard,
draft the script, edit, and rewrite it.

FURTHER READING
‘Computer Programming as an Educational
Tool in the English Classroom: A Preliminary
Study’, 2015, Global Engineering Education
Conference, Jesús Moreno-León and Gregorio
Robles (helloworld.cc/madrid).
‘Using Scratch to Teach and Learn English as a
Foreign Language in Elementary
School’, 2016, Sandra Costa, Anabela Gomes,
and Teresa Pessoa, International Journal of
Education and Learning Systems
(helloworld.cc/scratch).

helloworld.cc 47
 FEATURE

n Using robots in the classroom helps children engage with coding from a young age

CODING AND 21ST-CENTURY SKILLS

Ursula Martin shows how to include coding within a learning
framework that equips students for an uncharted future

ow should schools prepare students level and helps to prepare students for the room. This would require students
H for lives in a digital society with a digital careers. This preparation cannot using their critical thinking skills to
dizzying pace of change? Since the 1980s begin too early: it should begin in determine how to solve this problem.
educators, governments, charities, NGOs, elementary school and continue to develop
and companies have been trying to answer as students move on into high school. Coding and collaboration
this question. The 21st-century learning Collaboration is simply working with
framework, conceived in the US but applied Coding and critical thinking others to accomplish a specific goal. This
in numerous countries, attempts to equip Critical thinking is a skill that students may not seem too hard, but it is a skill that
students with the skills they will need. It need to develop in the classroom to help many students lack and truly need as they
moves away from rote learning towards an them to dispel misconceptions about what will eventually become a part of a society
engagement with higher-order skills. they are learning. This also helps students that functions best through collaborating
The four Cs of the learning framework to ask questions that will lead them to a with each other. Any student can learn to
are often used by educators in the better understanding of what they are code, but having students work together
US to guide their teaching. They are learning. Using coding to teach critical when coding opens a door to developing
collaboration, communication, creativity, thinking skills helps students to solve the better collaboration skills as well as
and critical thinking. As coding becomes problems they identify. For example, there enhancing their skills in the core subject
integral to how students use technology is a problem with lights being left on in areas. For example, students could
in the classroom, integrating coding into the classroom when the room is empty. increase their maths skills by learning to
the four Cs of the 21st-century learning Students could code an Arduino to turn code Dash and Dot robots to travel on a
framework takes education to the next the lights off if no motion is detected in specific path. The path could be designed

48 helloworld.cc
 n Students can enhance their
maths skills when learning code.
Coding and creativity
Creativity, as part of the learning process,
allows students to learn by thinking
outside the box. This helps students to
take ownership of their learning in a
non-traditional way. When coding
is added to this part of the learning
process, it introduces an opportunity
for students to use different types of
technology to ‘create’ what they are
learning. For example, as students study
biogeochemical cycles, they learn about
the carbon cycle and how it interchanges
with oxygen. A project on this could
include building a small greenhouse,
growing a plant in it and coding a
Raspberry Pi to open and close the
windows of the greenhouse as the oxygen
and carbon dioxide levels reach certain
levels. This level of creativity would help
students to understand just how these
biogeochemical cycles work.

A different way to learn

by groups of students working together, Framing educational activities using the
using basic geometry that the students four Cs of 21st-century learning challenges
would have learnt about in their maths students to learn in a non-traditional way.
classes. Having students collaborate Adding coding to this process changes the
to accomplish this task helps them to learnt. For example, if students are dynamics of learning altogether. Giving
understand that everyone’s thought tasked with coding a drone to fly, they students the opportunity to take ownership
process is not the same and that different must be able to work with others to of their learning will help teachers to
opinions can help to create a great project communicate exactly what they desire become facilitators and ensure that
and provide a richer outcome than one their drone to do (fly, flip, and so on) and students better understand how
person working alone. how long they want that action to take learning works.

INTEGRATING CODING INTO THE LEARNING

“ FRAMEWORK TAKES EDUCATION TO THE
NEXT LEVEL AND HELPS TO PREPARE
STUDENTS FOR DIGITAL CAREERS
Coding and communication place. This means using apps such as
Communication is key in education and the educational programming platform
in the workforce. Students must learn Tynker, to figure out how to accomplish
good communication skills to be effective this task. In this way, students not only URSULA MARTIN
in every aspect of their daily routine, learn a new ‘language’ but they also learn Ursula Martin has been an educator
whether that be in the classroom or in how to communicate what they have for 17 years. During that time, she has
their working lives. Coding introduces a learned by having the drone perform the taught Biology, Anatomy and Physiology,
different type of communication ability tasks they coded it to do. Communication and Environmental Science. She is a
into the learning process. As students is such an important skill for students to Raspberry Pi Certified Educator and
learn to code, they learn to communicate develop and effectively use, and acquiring District Level Technology Resource
by speaking and writing what they want coding knowledge can play a major part in Teacher in Mobile, Alabama
to create, or what they have previously learning this skill.

helloworld.cc 49
OPINION

EIRINI KOLAITI LEARNING MANAGER AT THE RASPBERRY PI FOUNDATION

IN DEFENCE OF PSEUDOCODE
Eirini Kolaiti explores whether describing code in simple English can
support the teaching and assessment of programming skills

The case for simple languages

W
hen I used to teach programming it didn’t take
me long to realise how easily my students could In 2006, Linda Mannila and colleagues from the Turku
confuse or even forget basic structural elements Centre for Computer Science in Finland compared 60
of a programming language. They would be tripped up by programs written by students aged 16–19 after their first
misusing capitalisation, white space, or parentheses. Basic programming course, in either Python (a simple language,
syntax errors like these would sidetrack students from developed for its readability) or Java (a more advanced
focusing on developing a solution. Instead, they would language). They analysed the assignments in terms of both
resort to trying out every possible instruction they could syntax and logic errors, and overall functionality.
find online in the hope that it would solve their errors. It The results were remarkable. Programming in Python
turns out that my experiences weren’t unique, and are not only helped students avoid making syntax errors,
also supported by research in computing education. but it also allowed them to solve the given problem. The
percentage of Python programs that ran correctly and
fulfilled the intended purpose was more than double the
Java programs.
One potential drawback of teaching programming in
Python rather than a more complex language, such as Java,
is that it does the students a disservice – that they would
need to relearn aspects of programming when moving to a
less intuitive language, like Java. However, the team did not
find this to be the case: students who had first learned with
Python were at no disadvantage when switching to Java.
So, if choosing a simple language can play such a
positive role in student progression, why not opt for
that option?

From simple to pseudocode

Allison Elliot Tew and Mark Guzdial took the matter of
programming language choice even further. In 2011,
they developed a way of comparing the knowledge of
university students who took introductory programming

50 helloworld.cc
courses in Java, Matlab, and Python. They wanted to make
the assessment language-independent, so they used
pseudocode instead of any of the taught programming PSEUDOCODE AT GCSE
languages. Pseudocode makes use of simple English to
describe what a program does. It is laid out in a similar REBECCA FRANKS
manner to a programming language, but removes some of Learning Manager at the Raspberry Pi Foundation

I recently asked GCSE Computer Science teachers on Twitter

A PSEUDOCODE-BASED
“
which type of pseudocode they used for designing programs.
38% followed the exam board specific pseudocode and 53%
ASSESSMENT CAN did not specify a syntax. Several commented that it was much
ACCURATELY DETERMINE more important that learners could effectively design their
own programs and the syntax used was less relevant.
STUDENTS’ COMPETENCY
the clutter that is needed for the machine to understand the
code, which increases the complexity to a human reader.
The results demonstrated that a pseudocode-
based assessment can accurately determine the
students’ competency regardless of their programming
background. This means that students could transfer
their comprehension of fundamental programming
concepts to pseudocode notation. Reversing this logic,
surely we can use pseudocode to scaffold the learning of There is currently no requirement for students to write
programming concepts. using a specific pseudocode syntax. However, most exam
boards have their own unique approach to pseudocode that
Walk before you run (a program) they use for presenting questions. Students will need to be
In 2004, an international group of researchers, led by familiar with these to answer the questions successfully.
Raymond Lister at the University of Technology in Sydney, Exam questions will indicate the form or response required,
conducted a study regarding programming competency whether it’s a specific programming language, natural
across seven countries. Instead of asking students to English, pseudocode, or flowchart.
produce their own programs, the researchers examined

helloworld.cc 51
 Linear search in pseudocode:

function linear_search(list, element)

for i = 0 to len(list) - 1
x = list[i]
if x == element then
return TRUE
endif
next i
return FALSE
endfunction

v = ['Bob', 'Doug', 'Alice']

value = input("Enter search string or q to quit: ")
WHILE value != ‘q’ AND value != ‘Q’
print( linear_search(v, value) )
value = input("Enter search string or q to quit: ")
ENDWHILE

Linear search in Python:

def linearSearch(list, element):

for x in list:
if x == element:
return True
return False
v = ['Bob', 'Doug', 'Alice']
while True:
value =input("Enter search string or q to quit: ")
if value.lower( ) == 'q':
break
else:
print linearSearch(v, value)

whether students could understand existing code (written

in Java or C++), for example, by predicting the output of a
given program. The results suggest that what stops many PSEUDOCODE ACTIVITIES
students from performing well in programming tasks is not
Here are some ideas for ways to incorporate pseudocode
a lack of ability to problem-solve, but a fragile knowledge of
into your teaching:
fundamental concepts. Students were unable to hand-trace
code (where the values of variables are calculated by hand)
n S tart activities with pseudocode to discuss concepts and
because of an inefficient command of basic programming
then convert into code to deal with syntax errors, and to
tasks, such as iterating over an array or the use of
test/debug the algorithm and check the structure of the
recursion. These areas are mostly related to an ability to
algorithm. It should be easier to write an algorithm after
read, rather than write, code.
students have worked on a pseudocode version, rather
I would therefore argue that as educators we should use starting with a programming language
programs in pseudocode to foster these preliminary skills, n A sk students to write a program using pseudocode
so that students benefit from a reduction in the cognitive and then swap with a partner for writing it in a
load caused by language-specific syntax. After all, most programming language
programming languages are not designed with the aim of n G ive snippets of pseudocode to test basic
teaching programming in mind, whereas pseudocode can misconceptions, for example in the use of recursion
be adapted to meet the needs of the students. n G ive small programs that students can hand-trace,
Pseudocode activities can be used in lessons in order to writing out the values of variables as the program
practise reading and tracing code, and as an opportunity progresses, to check their skills at understanding code
to discuss basic programming concepts. Students could

52 helloworld.cc
Linear search in Java:

import java.util.Vector;
public class JavaEx {
public static boolean linearSearch(Vector v, Object o) {
for (int i=0; i 5 v.size( ); i++) {
if (v.elementAt(i).equals(o)) {
return true;
}
}
return false;
}

public static void main(String args[ ]){

Vector v = new Vector( );
BufferedReader in = new BufferedReader(
new InputStreamReader (Systems.in)
);
String[ ] values = {"Bob", "Doug", "Alice"};
for (int i = 0; i 5 values.length; i++) {
v.addElement(values[i]);
}
String value;
while (true) {
System.out.println("Enter search string or" + " q to quit: ");
value = in.readLine( );
if (value.toLowerCase( ).equals("q")) {
break;
}
else {
System.out.println(linearSearch(v, value));
}
}
}
}

n Linear search in Python and Java from Linda Mannila , Mia Peltomäki, and Tapio Salakoski
then convert the pseudocode programs (their own or (2006) ‘What about a simple language? Analyzing the difficulties in learning to program’,
each other’s) into compilable code. This gives them Computer Science Education, 16(3), 211–227, DOI: 10.1080/08993400600912384 . Reprinted
by permission of the publisher (Taylor & Francis Ltd, www.tandfonline.com )
the opportunity to experiment with how to implement
algorithmic constructs using the specific features of a
EIRINI KOLAITI
programming language, and how to deal with syntax
Eirini is a Learning Manager at the Raspberry Pi Foundation.
errors. Having the stepping stone of pseudocode therefore
She is an RPi Certified Educator and used to be a Computer
helps with the skills needed to test and debug algorithms,
Science teacher. She misses her Girls’ Computing Club
Pseudocode can also be used for formative assessment.
(@eirinikolaiti).
Low-stakes tests and starter activities that use pseudocode
snippets to test specific misconceptions can help unpack

FURTHER READING
the underlying processes of program execution. I have
found that challenging students with small, targeted
pseudocode programs helps with engagement and
Mannila et al. (2006): helloworld.cc/simple
information retention. In this way, pseudocode could pave
the way towards gaining fundamental knowledge and skills Tew, A. E. and Guzdial, M. (2011):
through reading and tracing code before moving on to helloworld.cc/independent
writing actual programs and dealing with the inevitable Lister, R et al. (2004): helloworld.cc/study
corollary of syntax errors.

helloworld.cc 53
 FEATURE

CESI•CS: RECENT
DEVELOPMENTS IN
IRISH SCHOOLS
In October 2019, CESI (the Computers in Education Society of Ireland)
reviewed their Community of Practice (CoP) for computing educators

ince its foundation in 1973, the CESI•CS and CoP Improvements in computer
S
n

Computers in Education Society The CESI•CS CoP focuses on computing science education
of Ireland (CESI), has flourished as a education and is a collaboration with n Bottom-up collaboration at local levels

dynamic community of educators striving education centres within Ireland through and across sectors
to promote the use of computers in Irish the Education Support Centres of Ireland n Teacher readiness

education at all levels. It is a voluntary and Trinity College Dublin. CESI•CS began
grassroots organisation that draws its in September 2017 and was financially On Saturday 12 October, CESI•CS hosted
members from primary, secondary, and supported by Google for the first two years. its second National Symposium to bring
third level educators, as well as other The aims of the CESI•CS project are to make: together computing educators from across
relevant organisations. n A sustainable CoP for computer science Ireland to discuss the development of
in Ireland computing so far. A total of 50 computing

RELEVANT LINKS
n CESI – Computers in Education Society of
Ireland – cesi.ie
n CESI•CS – CESI community of practice for
computing – cesi.ie/cesi-cs/
n NCCA – Coding in Primary Schools Initiative –
helloworld.cc/primary
n JCTCIA – Junior Cycle Short Course Coding in
Action – helloworld.cc/aims

n CESI Logo @visualthinkery, licenced under CC-BY-SA

54 helloworld.cc
n Educators meet to discuss computing in Ireland thanks to CESI’s grassroots group

educators took part, representing all sectors, are keen to respond to the societal and CESI 2020 conference
levels and government agencies involved. parental pressure to introduce computing, A more recent highlight for Irish computing
The context of the CESI•CS symposium making a rationale for establishing the CoP educators was the annual CESI conference,
is that in the past few years, significant to prepare the ground. which attracted interest both nationally and
curriculum development work has occurred Since September 2017, CESI•CS has internationally. Impassioned computing
for computing in Irish schools: organised 57 regional and five national educators met on the day to share best
practices, innovations and research on all
topics related to the use of computers in

“ TEACHERS ARE KEEN TO RESPOND TO

THE SOCIETAL AND PARENTAL PRESSURE
education. The CESI 2020 conference
kicked off with a TeachMeet on Friday 28
February at 7pm in the Athlone Sheraton
TO INTRODUCE COMPUTING Hotel and continued all day in Athlone
Institute of Technology on Saturday 29
February. The theme of the conference was
n  or ages 12–15, there is now a computer
F face-to-face meetings all over Ireland ‘Our evolving learning landscape’.
science short course module, Junior Cycle in order to enhance the grassroots
Coding in Action. organisation of CESI based on trust,
n At senior cycle level, for ages 16–18, a respect, and collaboration. In addition, FURTHER
Leaving Certificate Computer Science
qualification equivalent to UK A-level was
educators could connect nationally by
signing up to the CESI mailing list. A
INFORMATION
introduced in 40 phase-1 schools from facilitator’s guide was distributed at the
September 2018. This will open up to a Symposium as a means to encourage and For more information about CESI, contact
wider number in September 2020. support volunteer educators to take over Adrienne Webb, CESI chair – [email protected]
n From September 2017, computational the organisation and facilitation of meetings.
thinking has been explored at primary A major focus of the discussion in
level through the Coding in Primary Dublin was the need for preparation of
Written by CESI members
Schools Initiative at the NCCA (National the second cohort of computer science
Council for Curriculum and Assessment) teachers, who will be teaching the
HELEN O’KELLY
which has involved some 40 schools over Leaving Certificate at senior cycle level DR RICHARD MILLWOOD
two years. from September 2020. CESI plans to ADRIENNE WEBB
continue support for them and others at KATIE MOLLOY
But not all schools and teachers can be a grassroots, regional, and national level, PAT SEAVER
part of these pilot programmes. Teachers both face-to-face and online.

helloworld.cc 55
 FEATURE

n Learners with more agency will be more engaged in their learning

PROJECT-BASED LEARNING:
A PATH TO AGENCY
Mark Calleja investigates project-based learning, a teaching method that
gives your students the tools to take charge of their own learning

roject-based learning is about Creating solutions from the ground up knowledge to other situations in their lives,
P asking your students to solve a Agency: For your learners, there’s a they understand the true purpose of the
real-world problem by designing and massive difference between being work they’re doing.
creating a project over a specific period of asked whether they know the answer Universal skills: Project-based learning
time. Getting started with project-based to a problem, and being asked to find a doesn’t give students ready answers to
learning in your classroom requires a bit solution to one. The first question assumes a specific problem; it asks them to build
of a shift in thinking: instead of providing knowledge (and thereby frames a lack of a mental toolkit for understanding any
an activity that demonstrates prior knowledge as failure) and has a narrow problem, so that they can create solutions
knowledge or acts as a unit assessment, focus, while the second gives learners the from the ground up. By enabling this in-
students’ learning is directly embedded room to be wrong or not know yet, and to depth learning, you equip your students for
into and emerges from the investigative develop real understanding and practical real life, letting them practise skills required
and design processes they engage in skills in a self-directed way. in most industries today: taking initiative,
while solving the problem you’ve posed. Engagement: Learners with more agency working responsibly, decomposing and
That is to say, their project is the unit: it’s are likely to be more engaged in their solving problems, collaborating in teams,
the vehicle for teaching the knowledge learning, too. A huge benefit of project- and communicating their ideas clearly.
and skills that students need to learn, based learning is the scope it gives you to
as well as the assessment process that set relevant, real-world problems for your Giving the power back to your learners
demonstrates their learning. I’m a huge students. Being able to relate their learning When your learners are interested and
proponent of teaching this way, for a to their own lives motivates students: when engaged with their own learning, your job
whole host of reasons. they see that they can apply new skills and changes from passing on knowledge and

56 helloworld.cc
 MARK CALLEJA
Resident trickster at the Raspberry Pi
Foundation and amateur gentleman
adventurer looking to go pro, Mr C is
a teacher, maker, hacker, MacGyver,
Raspberry Pi Certified Educator
n L earners need only teach themselves using the sheets
(@M1st3r_C).
and videos provided to make their ideas real

managing motivation to facilitation and have available. When they know what
PROJECT-BASED
inspiration. In project-based learning, you
direct learners towards information instead
functionality they want from their invention
(and what’s possible given their time and “ LEARNING ASKS
of handing them answers, and you support hardware constraints), learners need only
them in creating something they didn’t know teach themselves using the sheets and STUDENTS TO
they were capable of.
One way to get started with project-based
videos to make their ideas real. All of the
code is broken into three sections to make
BUILD A TOOLKIT
learning is to use a bank of Python resources each Python script modular; students can OF SOLUTIONS
that embody the idea, developed with the simply combine the code snippets on the
National Citizen Service. There’s a bank of sheets to make larger scripts that create
14 different helpsheets (with accompanying more complex functionality.
YouTube animations) that provides images
of each basic electronic component, a simple
If you’d like to replicate our hackathon
model with your own students, we have
RESOURCES TO
wiring diagram with numbered GPIO pins, released the facilitator’s guide to running SUPPORT PROJECT-
and the simplest of gpiozero code snippets the full two-day experience, complete with
to execute its basic functions, all on one session timings, delivery notes, workshop BASED LEARNING
handy page. slides, and a student support document
The helpsheets cover the most common called the ‘Developer’s Guide’ in which Guides for how to run hackathons with your
simple components used in digital making, participants can make notes and get students are available at rpf.io/hackathon-
from LEDs to infrared motion sensors, discussion prompts and tips throughout guides and rpf.io/component-sheets.
cameras, Bluetooth remote controls, and their build process.
beyond. There are also sheets that explain
the Sense HAT’s on-board sensors, joystick
and LED array, with accompanying code
examples. We also made some that cover
some commonly used processes in digital
making, such as playing sounds with Python,
making a remote control with the Blue Dot
Bluetooth app, and setting up a Raspberry
Pi-based gadget to function automatically
as your students intend it to as soon as they
power it up (known as ‘running headless’).
The intention behind the sheets is that
you will first support your students through
the design discussions they’ll need to
have before they start making things and
show them the library of components they n Why not run a hackathon with your own students?

helloworld.cc 57
 FEATURE

USING DEEPER LEARNING AS A

FRAMEWORK FOR TRANSFORMATIVE
EDUCATIONAL EXPERIENCES
Brent Richardson outlines the case for a pedagogical
approach that challenges students

benefit of completing school and Formal education continues to focus on eight categories, students rated their
A entering the workforce is being accountability tools that measure base level skills much higher than their prospective
able to kiss standardised tests goodbye. proficiencies instead of more complex skills employers had.
That is if you don’t count those occasional like problem solving and communication. Some people believe that this gap
‘prove you watched the webinar’ quizzes One of the main reasons the U.S. continues to widen because teaching
some supervisors require. education system is criticised for its within the confines of a standardised
In the real world, assessments often reliance on standardised tests is how test encourages teachers to narrow their
happen on the fly and are based on each this method of assessing a student’s instruction. The focus becomes preparing
employee’s ability to successfully complete comprehension of a subject can hinder students with a limited scope of learning
tasks and solve problems. It is often their ability to transfer knowledge from that is beneficial for testing.
obvious to an employer when their staff an existing situation to a new situation. With this approach to learning, it is
members are unprepared. The effect leaves students ill-prepared for possible that students can excel at test
higher education and the workforce. taking and still struggle with applying
A study conducted by the National knowledge in new ways. Educators need
Association of Colleges and Employers to have the support to not only prepare
found a significant gap between how students for tests but also develop ways
students felt about their abilities and their that will help their students connect to the
employer’s observations. In seven out of material in a meaningful manner.

BRENT RICHARDSON
Brent is the Fab Lab Director for the
community development organisation
BakerRipley in Houston, Texas. He studied Art
and Technology at the University of Oklahoma
where he used digital fabrication equipment to
create interactive gallery exhibitions.
n Outside of the classroom, assessments are often based
more on problem solving than standardised tests

58 helloworld.cc
n The project focused on capturing the
building of a new Fab Lab site in Houston

In an effort to boost the U.S. education

DEEPER LEARNING ENSURES THAT
system’s ability to increase the knowledge
and skills of students, many private “ LEARNERS ARE ACTIVE PARTICIPANTS IN
corporations and nonprofits directly support
public education. In 2010, the Hewlett THEIR EDUCATION
Foundation went so far as to develop a
framework called ‘deeper learning’ to help
guide its education partners in preparing they have learned and build upon that to powered time-lapse camera. Through
learners for success. create new knowledge. this collaborative project we quickly found
While deeper learning experiences are ourselves moving beyond classroom
The principles of deeper learning important for all students, research shows pedagogy to a ‘hero’s journey’ —
Deeper learning focuses on six key that schools which engage students from where students’ learning paths echo a
competencies: low-income families and students of colour centuries-old narrative arc in which a
in deeper learning have stronger academic protagonist goes on an adventure, makes
1. Master core academic content outcomes, better attendance and behaviour new friends, encounters roadblocks,
2. Think critically and solve and lower dropout rates. This results in overcomes adversity and returns home a
complex problems higher graduation rates, and higher rates changed person.
3. Work collaboratively of college attendance and perseverance In this spirit, we challenged the students
4. Communicate effectively than comparison schools serving similar with a simple objective, ‘Make a device
5. Learn how to learn students. This pedagogical approach is one to document the construction of Fab Lab
6. Develop academic mindsets we strive to embed in all our work at Fab Houston’. In just one sentence, participants
Lab Houston. understood enough to know where the
This framework ensures that learners finish line was without being told exactly
are active participants in their education. A deeper learning timelapse project how to get there. This shift in approach
Students are immersed in a challenging The importance of deeper learning was pushed students to ask questions as they
curriculum that requires them to seek out undeniable when a group of students attempted to understand constraints and
and acquire new information, apply what I worked with in Houston built a solar potential approaches.

helloworld.cc 59
 FEATURE

Students shared ideas ranging from

drone video to photography robots.
Together everyone began to break down
these big ideas into smaller parts and
n The project helped to develop the students’
better define the project we would tackle project management and engineering skills

together. To my surprise even the students

that typically refused to do most things
STUDENTS COULD UNDERSTAND THE
were excited to poke holes in unrealistic
ideas. It was decided, among other things, “ PROJECT ENOUGH TO BREAK IT DOWN INTO
that drones would be too expensive, robots
might not be waterproof and time was DOABLE PARTS AND DEVELOP A BUDGET
always a concern.
The decision was made to move forward
with the stationary time-lapse camera a remote power solution and building computer to dynamically turn images
because although the students didn’t know weatherproof housing. into video, one team member discovered
how to accomplish all the aspects of the A group of students found sample a new approach. Since the photos were
project, they could at least understand code for the Raspberry Pi that could be stored with a sequential numbering
the project enough to break it down into repurposed to take photos and store them system, thousands of photos could be
doable parts and develop a ballpark budget. sequentially on a USB drive. After quick loaded into Adobe Premiere Pro straight
Students formed teams and picked one success, a few ambitious learners started off the USB with the ‘Automate to
aspect of the project to tackle. The three working to automate the image post Sequence’ tool in Premiere.
subgroups focused on taking photos and processing into video. Eventually, after A great deal of time was spent
converting them to video, developing attempting multiple ways to program the measuring power consumption and
calculating solar panel and battery size.
n Students had to calculate the likely power
consumption and which battery size would be required
Since the project would be placed on a
pole in the middle of a construction site for
six months, the students were challenged
with making solar-powered time-lapse
camera as efficient as possible.
Waking the device after it was put to
sleep mode proved to be more difficult
than anticipated, so a hardware solution
was tested. The Raspberry Pi was
programmed to boot up after receiving
power, take a picture and shut itself
down. With the Raspberry Pi safely
shut down, a timer relay cut power for
ten minutes before returning power and
starting the cycle again.
Finally, a waterproof container had
to be built to house the electronics and
battery. To avoid overcomplicating the
process, the group sourced a plastic

60 helloworld.cc
 placed outside for a day to test the capability that allowed their camera to not only
of the device. survive but to capture a spectacular
The test appeared successful when the series of photos showing the impact
students checked the USB drive. The drive of the hurricane in the location it was
was full of high quality images captured placed. The footage can be seen at:
every ten minutes. When the drive was vimeo.com/238079409
placed back in the Raspberry Pi, a student
noticed all the parts inside the case moved. A worthwhile risk
The high temperature that day of Overcoming many hiccups throughout
the test melted the glue used to attach the project was a great illustration of how
everything. This unexpected problem the students learned how to learn and
challenged students to research a better to develop an academic mindset. A set
alternative and reattach the pieces. back that at the beginning of the project
Once the students felt confident in their might have seemed insurmountable was
device’s functionality, it was handed over to laughable in the end.
the construction crew to install the camera Throughout my experience as a
on a twenty foot pole. The installation classroom teacher, a museum educator
n A special box was created to house the camera and
Raspberry Pi, helping it to withstand stormy weather went smoothly and the students anxiously and now a director of a digital makerspace,
awaited to see the results. I’ve seen countless students struggle to
weatherproof ammunition storage box to Less than a week after the camera understand the relevance of learning and
modify. Students operated a 3D printer to went up, Houston was hit hard with the this has led me to develop a strong desire
create custom parts for the box. rains brought on by Hurricane Harvey. to expand the use of deeper learning.
After cutting a hole for the camera, a small The group was nervous to see if the Sometimes it feels like a risk to
piece of glass was attached to a 3D printed project they had constructed would facilitate learning rather than impart
hood ensuring no water entered the box. survive. However, when they saw that knowledge, but after seeing a student’s
On the rear of the box, they printed a part their camera had survived and was development into a changed person,
to hold and seal the cable from the solar working, they felt a great sense of pride. ready to help someone else learn, it
panel, where it entered the box. It only took They recognised that it was a makes it worth the effort. Let’s challenge
a few sessions before the group produced a collaborative effort of the group to ourselves as educators to help students
functioning prototype. The project was then problem solve possible challenges, acquire knowledge and use it.

n Work on the construction site was documented in

images, which were used to create a short video

helloworld.cc 61
 FEATURE

LEARNING GRAPHS: TOOLS

TO PLAN FOR PROGRESSION
Carrie Anne Philbin introduces an approach to map the computing curriculum

progression framework is the Firstly, there is the approach of creating

A backbone of any subject curriculum. a categorisation of skills and concepts into
It illustrates the sequence in which children a list or table. Sequencing is shown by
learn, noting where core understanding of a having objectives listed by key stage, year
topic is established in order to progress. As group or even by learners’ age. Examples of
part of the National Centre for Computing this approach include the CAS computing
Education (NCCE), my team is developing a progression pathways (helloworld.cc/
bank of teaching resources that deliver the path), and the Massachusetts Digital
computing curriculum in England. In order Literacy and Computer Science Curriculum
to successfully do this, we’ve studied a lot Framework (helloworld.cc/standards). They
of progression frameworks, examination are essentially a list of required knowledge
specifications, and even some research bundled by theme.
CARRIE ANNE PHILBIN papers. We found that there are two Another approach is to use a map of
Carrie Anne is Director of Educator quite different ways of presenting possible trajectories through learning
Support, Raspberry Pi Foundation, progression that show what should be waypoints—key building blocks of
leading the development of resources to taught and when it should be taught, learning—and how they connect to each
teach computing. as well as information on how or why other. This approach highlights where
concepts should be taught. prerequisite knowledge needs to be

62 helloworld.cc
 n The learning graph for the Year 8 topic of Python programming. Solid arrows show prerequisites while dotted arrows show a suggested order of learning

mastered before students can move on,

as well as the dependent knowledge
LEARNING GRAPHS DIRECTLY INFORM
contained in other ‘nodes’, each containing
one part of the computing curriculum “ LESSON PLANNING AND ALSO ADD VALUE
that needs to be mastered in order
to progress. Cambridge Mathematics BY SHOWING OTHER OPPORTUNITIES
(cambridgemaths.org) are leading
the way in ‘developing a flexible and
interconnected digital framework to help strand across an entire key stage, to the a lesson or unit. By checking that students
reimagine mathematics education 3–19’. learning objectives of a 6-lesson unit. are grasping the concepts, teachers are
We’ve been lucky enough to learn from When our team starts working on a unit, able to think more about how they are
their work, which has helped us to create the learning graphs are in a fluid state: teaching. They can revisit knowledge
learning graphs. they uncover the structure of the content that perhaps didn’t land with learners the
and the possible journeys through it, first time.
A tool for teachers without being bound to a specific teaching All progression frameworks are
The learning graphs organise computing pathway. The graphs eventually reach a subjective, and with little research into
content—concepts, knowledge, skills, and fixed state, where the nodes are further computing education, we rely on
objectives—into interconnected networks. structured and arranged to reflect our teachers’ experience of combining the
We found that nodes often form clusters suggestions on the order that the content ‘what’ we teach and ‘how’ to teach it to
corresponding to specific themes and we could actually be delivered. help inform this work. If you’ve not taken
could connect them if they represent two We believe that learning graphs could a look at our learning graphs, you can
adjacent waypoints in the learning process. be useful to teachers on a whole new level. access them via teachcomputing.org/
Depending on the context, the nodes in They directly inform lesson planning, but resources. Do let us know your thoughts
a learning graph could contain anything also add value by showing opportunities to by emailing us at resourcesfeedback@
ranging from the contents of a curriculum assess understanding at landmark points in raspberrypi.org

helloworld.cc 63
 FEATURE

BEHIND THE SCENES AT

AN ONLINE COURSE
Michael Conterio reveals some of the work that goes
into creating materials for online training courses

t’s been a busy year for my course comes first: working out what single person here helps cut down the time
I team at the Raspberry Pi it’s about. It sounds silly—surely when spent going back and forth on particular
Foundation—we’ve launched 14 online you start writing a course you know elements and can keep the course feeling
courses as part of the National Centre what it’s about—but even when you’ve more coherent. While writing the course
for Computing Education (NCCE). Here’s got an overarching topic you still need step by step, the lead educator tries to
how we’ve been going about making to work out the specifics to include. make sure that we’re not only sharing
sure that all of them are high-quality Sometimes, it’s a case of working out the useful pedagogical approaches, but also
learning experiences. programming concepts you want to cover, implementing them in the courses. We
the most sensible order to cover them know that teachers’ time is limited, so each
First, make a plan
Although we’ve launched an average of
more than one course a month, the story
behind each course starts five or six
“ IT’S A CASE OF WORKING OUT WHAT
PROGRAMMING CONCEPTS TO COVER AND
months before with the planning stage.
The most important part of creating a THE SENSIBLE ORDER TO COVER THEM IN

in, and some programs that the learners step needs to have a good reason to exist.
could work towards building. In other The flow of the course between steps
cases, the structure has been a lot harder is important too. We need to help keep
to pin down, for example in our ‘Impact learners feeling that the material makes
of Technology: How to Lead Classroom
Discussions’ course. In that case, we
had to work out how to make sure we
interleaved useful advice for running
classroom discussions and help for
long-answer questions with information
to start off discussions on a range of
different topics.
Once the overall content has been
decided on, we decompose the course
down into smaller segments. One person
then takes control to plan each course step
n Courses use screencasts so that the instructors
can talk you through writing code. by step. Dropping down from a group to a n Each course has one person overviewing each step

64 helloworld.cc
n Collaboration is an important part of course design

sense and motivates them to continue. several months, thanks to a combination as trying more generally to minimise
Not getting this right at this stage can of factors, such as the amount of material confusion. They’ve also got to try and
lead to trouble. Although we may make needed (courses can have over 25,000 make sure that there’s enough of their
minor changes to the plan as we write the words), the fact that our writers are often own passion and voice present to make
course, the few times we’ve made larger juggling multiple responsibilities, and also a connection with each of the potentially
changes, we’ve had to put in a lot of extra because I can get quite picky. In my role thousands of learners per run. It’s not an
work to make sure the learner’s journey as production manager, I see the initial easy job, but they all put in a lot of hard
through the course still makes sense. drafts of each step and generally provide work, and take a lot of feedback with
Before the writing can start, the step- feedback. At this point I’m pretty much good grace, to make the courses the best
by-step plan will be checked by others in acting as an advocate for the course that they can be.
the team, particularly those who helped participants, trying to make sure that
with some of the earlier planning. This everything is as clear as it can be for
can help spot where some elements from them, whether that’s a definition of a term,
the initial planning haven’t been followed instructions for activities, or in making
through in the step by step, as well as sure that learners not only understand
point out a few points where the plan can what they are doing but also why we are
be polished up slightly. asking them to do it. Sadly, I am fallible,
but other people will also get a chance to
The write stuff check the course later and we can also
Once the plan is approved, it’s time to respond to feedback from our learners.
get down to the writing. This can take Writing for online courses is pretty
hard on our creators, as
it’s quite different to the
writing (and teaching!)
they usually do. Although MICHAEL CONTERIO
learners will be able to Michael works for the Raspberry Pi
comment and get feedback, Foundation, producing online courses.
real-time communication Previously a physicist, he moved into
isn’t possible, so our course education and helped develop Isaac
creators must navigate the Physics. In his spare time he enjoys
tricky task of conveying walking, board gaming and performing
tone in solely text-based stand-up comedy.
n Real-world examples, like creating a train departure board,
help teachers make programming relevant to students communication, as well

helloworld.cc 65
 n All our teams work together to ensure good high-quality learning materials

Greater than the sum of its parts

As the course launch date nears, we have
ONLINE COURSES FOR COMPUTING EDUCATION to start putting all of this material together.
Our team must adjust animations to match
There are over 20 free online courses from the courses as credit towards your NCCE Computer
the voiceover timing and edit them into
Raspberry Pi Foundation, available to access from Science Accelerator Programme certificate - a
the videos. Other members of the team
anywhere in the world. Whether you’re simply development programme with a mixture of online
must send the videos off for transcribing
curious about computing, a coding club volunteer, and face-to-face modules to help you confidently
and then check the transcriptions are
a lifelong learner, or an school teacher, we’ve got all teach GCSE computer science. Visit the NCCE
accurate. We also need to upload the
your learning needs and interests covered. website to find out which courses are part of
images and write suitable alt text for
And if you’re a teacher in England, you can use a this programme, and book your place today at
them, for accessibility. Our wonderful
subset of the Raspberry Pi Foundation’s online teachcomputing.org.
copy editing team check through the text

Illuminating computing managers to ensure that all of these pieces

“
One part where we’re very lucky is with our of media come out looking as close as
illustrators, animators and videographers. possible to how they were imagined. PRESENTING IS
They all work together with the course writers Our lead educators often have to VERY DIFFERENT
to not only make the course look good but step out of their comfort zone at this
to fill it with memorable content that aids point, as they find themselves in front FROM BEING IN
the learners comprehension of the material of a camera. Presenting like this is very THE CLASSROOM
covered. While writing the course, the author different to standing in front of a class
adds notes on content they would like added – looking directly into a camera can feel
to their text which they later discuss with very strange, especially when there’s for the inevitable typos and unfortunate
the other content creators. The production of an autocue to read. The scripts loaded phrasings, and fix them to make sure that
these assets is a big job and can take up to onto the autocue have been adapted the meaning isn’t lost. And finally we’ve
two months for each course, despite having from the course text by one of our video got to upload all of these things onto the
different people working on each aspect. As production managers to cut them down online platform, using a seemingly never-
the people producing these are not subject for time, as well as to adapt anything ending array of text boxes and buttons. At
specialists, it’s easy for a key part of an that sounds more awkward when this point both us and the web team make
illustration to be misinterpreted or part of an spoken rather than written. In any case, a final check, and the course is done!
animation to accidentally pave the way for our videographers guide our educators Now all we’ve got to do is facilitate the
a misconception. To avoid this, all of these through the whole process and by the course, responding to discussions, queries
people have to continue to work closely end some find that they even enjoy being and cries for help, until the course run
with the course author and the production in front of a camera. comes to an end. Well, until next time.

66 helloworld.cc
 LESSON PLAN

AGE RANGE
14–16 years
PRACTISE
WITH POETRY
YEAR GROUP
UK years 10
and 11

LESSON TYPE
Unplugged (pen Adrienne Tough shares an approach to support
and paper)
students in their use of computing vocabulary
REQUIREMENTS
tudents can find some The students are presented with and helps develop their reading and
• Poem to annotate S computing concepts quite poems from a range of topics. They comprehension. The activity develops
• Pen and paper intimidating – particularly the can choose which one to focus on, or key literacy skills, and seems to be
extended vocabulary. So I’ve been the teacher can decide. The poems well received. This method not only
keen to find creative ways to make can be effective in a revision lesson helps develop oracy – an invaluable
them more accessible. One method before an assessment. The students skill – it also keeps the students
I’ve developed is writing poetry can recall the key concepts taught in engaged as they wait to see who
for the classes – and eventually, I’d the module and annotate the poem will be selected, and the rhyming
love them to write their own, too! to check how much they have learnt. helps make the content memorable.

DIFFERENTIATION POETRY CAN BE A CREATIVE WAY

Students with SEND can be supported in this
“ OF MAKING COMPUTING CONCEPTS
activity by grouping students together and giving
them more support with annotations. A keyword
MORE ACCESSIBLE TO STUDENTS
bank can also be provided so that they can identify
the key concepts without having to read the whole So far, the poems have been Students have even asked for extra
poem, if that is easier. used as a classroom-based activity, copies to annotate and some have
To challenge high attainers, you can encourage as well as being given to students quoted lines back as an answer to a
them to write their own poem to contribute, or you as optional revision. If I know the question posed in a later lesson.
can set high expectations of their annotations.
answer is in a poem, I will direct When introducing poems, I have
For students with English as an additional
them to read the relevant poem and found it has been helpful to model
language the poems can easily be translated
online, and images can be paired to help explain then turn the question back on the expectations and demonstrate how to
the concepts. students. This makes them more annotate a poem before setting
responsible for their own learning students the task.

ADRIENNE
TOUGH OBJECTIVES
Adrienne is a Computing
teacher in West London.
She recently completed To recap different computing concepts (depending
the Teach First graduate on the poem), particularly targeting common
scheme. She is keen to
misconceptions
investigate and deliver
creative measures to
help students understand To recap and define key words associated with the
key computing concepts unit that has recently been covered
(@tough_miss).

68 helloworld.cc
STARTER ACTIVITY:
INTRODUCTIONS 5 MINUTES
Students will read the poems independently and become familiar An optional activity for higher attainers is to write down the
with the content. I am keen to save resources, so rather than additional information that the poem has helped to generate.
printing them out, I email them to the students and present them
on the board. I also like to print out some spare copies, as some
students will want to draw on them.

ACTIVITY 1: READ ALOUD 5 MINUTES

Display the poem on the board and have the students each read concepts that are introduced in the poem.
a couplet (that’s a pair of lines), then choose someone else to Circulate around the room to keep the students on task and to
continue reading from there. Otherwise, you can pick students to highlight key concepts or help with things the learners stumble
read. If the students choose to add a rhythm, encourage it! The on. Support students to help them use the correct pronunciation,
more fun they have, the more they are likely to remember the and read together as a class at least a few times.

MAIN COMPONENTS
A POEM ABOUT HARDWARE
Hippos is how we remember the components of our PC
It really is that easy, just read on and see
H is for hardware, the things that are physical
I is for input devices, entering things so they’re digital
P is for primary memory, R starts our main two
P is for the processor, also known as the CPU
O is for output, so that we can hear and see
S is for secondary storage; our files are not just temporary
Software, of course, is needed too
Our programs and apps give us so much to do!

helloworld.cc 69
 LESSON PLAN

PRIMARY MEMORY: A POEM ABOUT ROM AND RAM

Directly accessed by the CPU,

That’s our primary memory, remember the main two:
First there’s RAM, temporarily storing what we’re on
But forget to click save and then it’s gone!
Being volatile means to save it needs electricity,
It also allows you to access and edit easily.
Next we have ROM, this enables our computer to start
But to access and edit we need to be very, very smart
Is it volatile? No, it’s not,
Our contents will not be forgot!
There are others of course, like cache and virtual,
So keep reading on to become even more intellectual!

ACTIVITY 2: MAIN ACTIVITY – RIP IT APART!

Students will now annotate what you have written, so encourage and embrace their
critical eye. You could look at what has been highlighted if this was completed in
previous step, or highlight the key concepts now. Perhaps ask a student what they
would highlight. Students will then be questioned on the content of the poem.
THE ACTIVITY HELPS
“
For example, in one poem on computing components, I have written:
“P is for primary memory, R starts our main two.”
TO DEVELOP KEY I then ask what R stands for. If anyone doesn’t know, they have to try to find
LITERACY SKILLS out. I then ask for further elaboration – for example, “What other characteristics do
you know of ROM and RAM? What other examples of primary memory could you
give?” Pose a few questions and then leave the students to continue to analyse

RELEVANT LINKS independently, with partners, or in groups. Circulate and assist where necessary.
Students can analyse the poems on printouts or write in their books. They
Adrienne’s poems are available at could create mind maps or flash cards based on the reading, so they are also
computingpoetry.weebly.com. creating their own revision material. Students often do not know how to revise, so
going through these methods can benefit them, not only in computing, but in their
‘Computational Fairytales’ by Jeremy
Kubica helped inspire Adrienne’s idea: other subjects too! If working in groups, work or groupings can be swapped and
computationaltales.blogspot.com. other poems can also be distributed.

ASSESSMENT
ACTIVITY 3: PLENARY ACTIVITY –
This can work as an assessment or as a starter WHAT’S BEEN READ SHOULD BE SHARED 5 MINUTES
for the next activity: present the poem again,
but with gaps in, then ask the students to
Ensure students have had the opportunity to work with others, and that they
rewrite and complete the missed content
have shared their ideas and looked at any other creative projects that have been
This lesson assists revision for end-of-topic inspired to get started with.
assessments. Short quizzes or games such Towards the end of the lesson, choose students to recall what they learnt from
as Kahoot! can be used to ask questions the poem – without looking, if they can. Have the other students elaborate or
about what the poem is covering. challenge them as necessary.

70 helloworld.cc
 LESSON PLAN

AGE RANGE
11­— 13 years
CREATE A
HOLOGRAPHIC
YEAR GROUP
Year 9

LESSON TYPE

ILLUSION
Text-based
programming

REQUIREMENTS
• p5.js library
Nazia Fakhruddin introduces a project in which students explore how 3D
• p5.js web editor
• 3D holographic programming and optics can be used together to create awe-inspiring results
pyramid or
hologram is an image
clear plastic A that appears to be three-
water bottle
dimensional and can be seen by
the naked eye from all sides. A real
hologram can be seen without the
use of any special lenses or screens;
images that appear to be holograms,
but are actually produced by placing
lenses in a certain way to create the
illusion of a three-dimensional ghostly
image, are not true holograms; these
are called holographic illusions. They
were used for the first time to create a
theatrical illusion, known as Pepper’s For this project, we will be creating
Ghost, by an inventor called John 3D objects using p5.js WEBGL library

OBJECTIVES
Henry Pepper in 1860. and a 3D holographic pyramid, which
When I first came across this will act as a lens, to observe our
illusion, I thought it was both really holographic illusion. Inexpensive 3D
interesting and easily programmable holographic projection pyramids can
using p5.js. I introduced 3D be bought online, or you can use the 3D shapes
programming concepts to my class top of a clear water bottle. When an
using this project and the students image from a screen is reflected at an Pepper’s ghost
enjoyed every step; they also went on angle of 45 degrees onto a
to explore and experiment with ideas transparent screen, a reflected virtual
pyramid principle
of their own. 3D image with depth is created.

WHAT IS P5.JS? DIFFERENTIATION

p5.js is a JavaScript library for creative coding, with Support: Stretch and challenge:
a focus on making coding accessible and inclusive for Get the students to create Make two shapes and try
artists, designers, educators, beginners, and everyone the different 3D shapes and rotating along different axes
else. It is free and open-source. rotate them. Ask them to give by using the rotateX, rotateY,
3D colour to the shapes. and rotateZ functions. Can
For more information on p5.js, see Saber Khan’s feature your students create a rotating
on pages 44–45 of Hello World issue 8. Saturn together with rings?

helloworld.cc 71
 LESSON PLAN

ACTIVITY 1:
function setup(){
Let’s start by creating a new project using the p5.js web createCanvas(600,600,WEBGL);
editor. You can find it at rpf.io/p5js. }
It has the following two lifecycle functions: the function draw(){
setup() function, which is called only once at the start of background (0);
the program execution, and the draw() function, which } 1
is called repeatedly, clearing the canvas every time.
Edit the setup() function, by changing the canvas
dimensions and adding the parameter of WEBGL in
createCanvas().
In the draw() function, change the background() to
0, as we need a black background so we can view our
illusion clearly. 1

In the draw() function, we need to make four boxes in

four positions: up, down, left, and right.
Start by creating the first box between push() and
pop() functions. The box will appear in the centre of the
stage when you press the play button. 2

To change the placement of the box, it needs to be function draw(){

translated. To do this, use the translate() function, keeping background(0);
the x position as 0 and changing the y position to -200. push();
The box should appear at the top of the canvas now. 3 box(50);
pop();
} 2 function draw(){
//top
Try challenging your students by then asking them to function draw(){ push();
add and translate three more boxes in a similar way. background(0); translate(0,-200);
Each new box should be between its own push() and push(); box(50);
pop() function. Here’s what they might come up with. 4 translate(0,-200); pop();
box(50);
pop(); //bottom
} 3 push();
translate(0,200);
box(50);
pop();

//right
push();
translate(200,0);
box(50);
pop();

//left
push();
translate(-200,0);
box(50);
pop();
} 4

72 helloworld.cc
ACTIVITY 2:
Start by making a global variable called angle and initiate it with 0 at the top of the script.

angle=0;

Then, in the function draw(), increment the angle by 0.003, so the angle changes slowly.

function draw(){
background(0);
angle+=0.003;
push();
...

Now, in the function draw(), on the line below the first translate(), add a rotateX(angle).
When you press play, the box should now be rotating on the x axis. Try adding
rotateY(angle) and rotateZ(angle) functions as well. Have your students try and make
all the boxes rotate in the same way. NAZIA FAKHRUDDIN
Nazia is a Royal Institution
push(); Computer Science Masterclass
translate(0,-200); Speaker and has been a Code Club
rotateX(angle) UK Volunteer for the past six years
rotateY(angle) in the local library in the London
rotateZ(angle) Borough of Hounslow; she is the
box(50); author of the book Superfun p5.js
pop() Projects (@NZAmbience) .

ACTIVITY 3:
To give 3D colour, use the function normalMaterial() at
the top of the draw() function. Also use stroke(‘yellow’)
and strokeWeight(3), to highlight the box boundaries.

function draw(){
background(0);
angle+=0.003;
normalMaterial();
stroke(‘yellow’);
strokeWeight(3);

Place the 3D holographic pyramid or water bottle in the

middle space between the boxes. Now observe from
the side the holographic illusion of a rotating cube.
The complete code for this project can be found at
rpf.io/p5js-hologram.

n If you don’t have a 3D

holographic pyramid
you can also use a
clear water bottle to
see the holographic
image, which will be
small but visible

helloworld.cc 73
 AGE RANGE
11–13 years GETTING STARTED
WITH PYTHON
YEAR GROUP
Australia Grade
7, 8 and 9

LESSON TYPE
Tim Bateup shares his own experiences
Unplugged
of introducing students to Python
OBJECTIVES
text-based
programming
fter being told that my basic coding concepts and develop
A
REQUIREMENTS department had no room their understanding. Once they
in the budget for flashy classroom have watched a few tutorials, they To understand
• A computer with
Mu installed –
resources, I needed to come up then try their hand at completing how computer
with a way to teach coding in an some practical activities to test their
codewith.mu
engaging manner that didn’t cost new-found knowledge. In total, the
programs work
• A cup, a spoon, a a cent. As a result, I designed a workbook contains 24 activities for
jar of chocolate Python workbook, specifically students to complete. To be able to write
milk powder, a designed for teachers who are Though the video tutorials and
carton of milk
algorithms and
looking for a completely free way associated activities are not covered
to teach coding to beginners. in this article, I focus on how represent these
This workbook starts out with a algorithms can be introduced to the diagrammatically
basic introduction to what computer students early in the term in a hands-
programs are and how they work. on and practical way. Students
as flowcharts
Students briefly look at algorithms usually start off quiet in this lesson,
and flow charts before jumping but once they start acting out the To be able to write
straight into their first experience algorithms, the whole classroom
simple programs
of coding in Python. As the term comes to life with lots of laughter
progresses, students watch just 21 and plenty of learning using Python code
short video tutorials that teach them and engagement.

ACTIVITY 1: THE FIVE-MINUTE CHALLENGE 5 MINUTES

At the start of every lesson, as students In today’s lesson, the five-minute robot to complete this task, the student
come into the room and get themselves challenge is for students to write out the must give the robot clear and easy-to-
sorted, display a question on the projector step-by-step instructions for making a follow instructions (an algorithm).
related to some coding completed in glass of Milo—a chocolate and malt drink The students are given five minutes to
previous lessons or something that will be popular in Australia. write down the step-by-step procedure for
covered in that particular lesson. Students need to imagine that they are the robot to make a glass of Milo. Students
The question could be getting students living in the near future and hanging out will need the instructions they have
to draw a flow chart, adding comments at home with their pet robot. They start to written for the next part of the lesson.
to some prewritten code, or even finding feel a bit thirsty so they tell their pet robot
errors in code and fixing them. The first to make a glass of Milo for them. For the
two students to solve the problem each
receive an award, for example, I brought
STUDENTS MUST
in Mario and Luigi hats that the successful
students could wear for a day. The kids “ GIVE CLEAR
love to wear them, and they make the
classroom environment a little more of a INSTRUCTIONS
positive place to work in.

74 helloworld.cc
ACTIVITY 2: ACTING OUT THE ALGORITHM 40 MINUTES

When the students have their algorithms for n How could the algorithm be improved?
making a glass of Milo, they will complete n Could certain steps have been completed in
each other’s instructions. It’s a good idea to a more efficient manner?
have some towels on the ground or to move
the class outside at this point to avoid making Repeat the acting and debriefing for two more
a mess in the classroom.
Ask for volunteers to act the role of a robot.
algorithms with different volunteers. Once
finished, bring the class back together and
ASSESSMENT
Have your robot stand out in front with the explain to students how today’s lesson is an
To assess this unit of work, I like
milk, Milo, cup, and spoon on a table in front analogy of how computers read instructions to run two assessment tasks:
of them. Ask for another volunteer to read out and process code. Students will then begin 1. An exam that is made up
their algorithm, one step at a time. The robot to understand the importance of breaking of a theory and a practical
must act out exactly what they hear for each down problems into small chunks and creating component.
step and not let any human instinct take over. specific instructions for the computer. 2. A small project, where
This usually results in some hilarious outcomes Moving on from here, students continue on students work in pairs to
(and a bit of a mess). through the workbook to look at flow charts solve a number of real-world
problems using Python code.
After the first algorithm is complete, conduct a and learn how to represent their algorithms
These assessment tasks will
quick debrief with the class. Ask questions like: diagrammatically. They then dive right into
give you a clear idea of the
n What steps worked well? coding where they can see first-hand how a capability of each student, while
n What steps didn’t work so well? computer processes instructions. also lending themselves to a
range of 21st-century skills such
as critical and creative thinking,
communication, collaboration
and teamwork, and ICT skills.

TIM BATEUP
Tim Bateup is a Digital Technologies
teacher at Noosa District State High
n Mu is a simple Python code editor, useful for students starting to code. School in Queensland, Australia.
He has been teaching for 13 years
across a wide range of subject
DIFFERENTIATION areas. He has a YouTube channel:
www.youtube.com/tbate54.
The use of video tutorials throughout this unit of work allows students to work at their own pace,
in a ‘flipped classroom’ setting. Students are encouraged to complete the video tutorials in their
own time, either at home or during lunch breaks. They then come to class with any questions
they might have and will generally spend the lesson completing activities in the Python
RELEVANT LINKS
workbook. While this approach does require time management on the part of the student, they
generally do a good job of this once they are shown how. Workbook Download
With video tutorials, students put headphones on and seem to enter their own world – there are helloworld.cc/workbook
very few distractions in the classroom and most students work quietly for the entire lesson.
It is common to see students refer back to the video tutorials throughout the course as they Teaching Python Podcast
complete the activities. This helps them refresh their memory on how to complete certain tasks
and saves the teacher from having to repeat themselves. helloworld.cc/podcast

helloworld.cc 75
 LESSON PLAN

AGE RANGE
7–10 years
MAX: A MACHINE
LEARNING ADVENTURE
YEAR GROUP
Greek Year 9

LESSON TYPE
Unplugged (pen
Machine Learning core ideas and concepts can be presented to children like a game
and paper) and
presentation
f you can’t explain it simply, example, if we provide images that
‘I you don’t understand it are dark and blurry, then the model
REQUIREMENTS
well enough.’ In this famous quote, would not learn from them. OBJECTIVES
• Laptop, HDMI Albert Einstein suggest that there In feature extraction, we need
cable, large screen is no such thing as a subject so to detect the useful information
• Whiteboard and difficult or complex that we cannot that is hidden inside the data, Discover machine
markers explain it to children. which will be used later in machine learning
• Pen and paper This is the challenge we undertook learning algorithms. A better way
• Combination lock, last May when we set out to explain to summarise that would be ‘find
Learn some uses of
chestandchocolate core machine learning concepts those characteristics which best
coins (bonus step) to children, in the form of a game. define your data’. Feature extraction machine learning
For the most part of the session, is performed by the user manually
at our CoderDojo in Votanikos, a and is not part of a machine Understand
neighbourhood in Athens, Greece, learning algorithm.
we used presentation slides. At the The next important issue is and apply basic
end of each task, the learners had the selection and training of the algorithms of
to find the same answer to that of model. This is again strongly
machine learning
our robot, Max, in order to succeed. related to the nature of both the
We did not use any coding or problem to address and the data. (no coding)
engineering parts, since our ‘robot’ For example, if you want to predict
was an avatar. We had eight children a yes/no task, your problem is
aged 8–10 attending the club at defined as classification. If you stories that involve children as an
the time. know your labels beforehand, this is essential part of that story, and be
supervised classification. more careful with concepts where
Before the black box higher levels of mathematical ability
A common perception of machine A lasting impression are required, where our the approach
learning is that it works like a black Six months after running this needs to be adapted. But we are on
box, or like magic: that it receives activity, we asked the children the right path.
data from one point and produces a number of questions about If you find this adventure as
an answer at the other, and it is this session. interesting as we did, then you can
up to the algorithm to interpret the The story and its critical details try making your own sessions based
data in an appropriate manner in were mostly remembered. Core on this material. Children will have
order for the model and reach a machine learning concepts like the fun experiencing their first contact
decision. While this is partly true, features and the decision trees with machine learning concepts,
there is a very important stage, were recalled, whereas there were getting prepared for a digital world
before the use of each model, which some difficulties with the ‘k-Nearest where these concepts will most
we call data preprocessing and Neighbours’ example. The children likely have an important role. You
feature extraction. enjoyed their reward at the end. can contact us via email at
In data preprocessing, we use This gives us confidence that [email protected] and ezeakis@
various ETL (Extract, Transform and we can organise more lessons on gmail.com, with any questions. We
Load) methods to process the data difficult subjects in the future. We will be more than happy to help you
and clean it as much as possible. For have learned to focus more on where we can!

76 helloworld.cc
ACTIVITY 1: SETTING THE SCENE:
MAX THE PIRATE-ROBOT 5 MINUTES
We started the series of activities by introducing Max:

‘Max is a robot that has a dream of becoming a pirate and finding treasure. Since he
doesn’t know a lot about the world, we need to help him. We will help him find a ship, a
crew, a map, the treasure island on the map and then to find the treasure on the island.’

ACTIVITY 2: MAX FINDS A SHIP:

THINK LIKE A COMPUTER 5 MINUTES
‘Max needs to start his journey by finding a ship. But he doesn’t know what a ship is or
what one looks like. We need to find a way for him to understand what a ship is.’

Our learners need to understand that there are no obvious answers when it comes to
computers. We need to provide data to computers in order for them to understand. Max has
no idea of the concept ‘ship’ except that he needs one. We asked our learners what would
be the easiest way for Max to learn about ships and they all agreed to show him images.
Therefore, to start with, we provided him with images of ships, so he could understand what
a ship looks like.

ACTIVITY 3: MAX FINDS

A CREW WITH FEATURES 15 MINUTES
‘Max has to find a crew for his new ship. There are many people he can choose from,
but he has to be smart. If he chooses two similar people, then they may team up in the
future and cause a mutiny onboard. How can he learn things about them, that will help
him see if they are similar or dissimilar?’

In this step, the mission was for Max to find a crew for his new ship. This discussion
introduces what features are and why they are important for the solution of each problem.
We showed our learners specific images of people and asked them questions like: What ALEXANDROS
makes each picture different? Why is Pirate 1 different to Pirate 2? And finally, is there a
ZEAKIS AND
global good feature that can always be used to resolve any given problem?
The answer to the last question is no and the children saw this. For example, if you are
MANOS ZEAKIS
Alexandros Zeakis is a Data
looking for a picture of a child, then you would discount any bearded people, but not all non-
Analyst in a research centre. He is
bearded people are children. So, we asked them to find the two most dissimilar people in
interested in inspiring young people
the photos. We used eight predefined features (e.g. hair length, beard, eye colour, etc.) and
within Data Science.
then found the two images which were least similar. We identified that two dissimilar people
Manos Zeakis is a DevOps
would probably not team up and cause a mutiny onboard. A symmetrical task would be to
Automation Engineer. He volunteers
find the two most similar people with the notion that they would collaborate more easily.
as a Dojo Champion and Raspberry
Therefore, each problem requires a different approach in the features identified.
Pi Foundation Translator.

helloworld.cc 77
 LESSON PLAN

ACTIVITY 4: MAX FINDS THE GUIDE: DECISION TREES 15 MINUTES

‘Max has to go to the market to find a specific person that will give him
the map. There are other pirates in the market as well, that will try and Bearded
trick him, so he has to learn beforehand how to identify the correct
person. He has already learnt about features, so is there a way for him to Y N
utilise that knowledge and find his guide with the correct questions?’
Long
Here, we introduce our first machine learning model — a Decision Tree. It’s a Earring
Hair
relatively easy model to explain as it uses words rather than maths, and can
be drawn. The general idea is that you start from the first question (root) and Y N Y N
each answer leads to a different question (branch). If the tree is built correctly,
the last question should lead to your answer (leaf). The game Guess Who?
works this way — ask the right questions and based on the answers, you Pirate 1 Pirate 2 Pirate 3 Pirate 4
will reach the right person. Again, by using specific images of people, we
designed our own Decision Trees based on features that our learners detected
n DecisionTreesresemblethegameGuessWho?.Maxusesthisapproachtofindthe
in the images. Please note, there is no one correct answer! person with the map.

ACTIVITY 5: MAX FINDS A MAP

15 MINUTES
RULE 1

A ship can move up,

down, left or right

RULE 2

‘Max visits the area on the map, only to find that a war rages
between two major pirates that want to conquer the same
islands Max wants to. He could wait for the war to be over and
then find any unoccupied island, but he is impatient and wants A ship can occupy an

ASSESSMENT to start anyway. Based on the ancient code of pirates regarding island if it is adjacent
to it and there is no
naval battles and his wit, Max will try to work out the outcome castle

of the battle using these rules.”

There are two ways to assess the
learners throughout the lesson.
RULE 3
Another machine learning model that can be easily explained in
The competitive way would be
to split the learners into groups natural language is ‘k-Nearest Neighbours’. When a predefined
and have them solve the puzzles distance exists on labeled data, a new unforeseen piece of data
in each task as a group. Each can be labelled in relation to this previous data (hence the ‘Nearest
group that solves the puzzle Neighbours’ part). The main difference with Decision Trees, that we A ship can occupy an
island with a castle
first takes the most points, the used above, is that Nearest Neighbours requires maths. This is the if there are two
second group one less, etc. occupied islands
main reason that the exact form of this model was not used and we adjacent to it
The collaborative way would be adapted the model to a game so that the learners could understand
to have the children solve each
the main idea. The game had three simple rules (see image) and
puzzle as one group.
then the learners predicted each pirate’s move in the map, based on the given rules and the
We preferred the benefits for them. The final state of the map after the battles exposed a single island that
collaborative way.
Max could safely visit.

78 helloworld.cc
n Initial unsolved puzzle

DIFFERENTIATION
ACTIVITY 6: MAX FINDS THE TREASURE:
LANGUAGE MODELS 15 MINUTES Since the course material was
mostly visual, some adaptation
would be needed for those with
‘Max enters the cave on his island and finds a chest that is locked. This particular lock vision impairment. For example,
requires no key, but a numerical combination. His only hint is a piece of paper next to in the activity where students
his chest, which includes a small phrase with a riddle at the end. How can he use this identify features, you could
phrase to find the answer?’ change to looking for a high-
pitched voice.
Next, we used language models from text analytics. The main idea is that when you write If you have learners with SEN in
a new text, you don’t always produce new context but rewrite something you have written your group, modify the pace of
the course to suit their needs.
previously. If someone has an old written text of yours and you begin a new phrase,
Some learners get very excited
then they will most likely be able to predict what you will say. This is a theory based on and rush through the puzzles
probabilities, but we simplified this by looking at the frequency of words used when the only while others may need extra
information was one letter. For instance, if you have the phrase ‘This is an item’ and someone time, so ensure everyone is at the
gives you ‘i_’, you will think that ‘s’ will most likely be the next letter, since ‘is’ occurs twice same stage.
and ‘it’ only once. Following these rules, Max used the given text (see image) and found the With advanced learners, instead
secret combination to unlock the chest, claim his treasure and achieve his dream of becoming of making the questions more
a pirate.
difficult, which could alienate
them from the rest of the group,
As a bonus, Max offered our learners a second chest — physical this time — with a second ask them to be assistants during
phrase, where each team of two learners would have to find the correct digit. When all the presentation. That way,
teams found the correct digits and unlocked the chest, they found a treasure of chocolate they would remain engaged
coins inside. throughout the course and
prove that they understood
each concept well enough to
explain it easily to others. An
extension of this lesson would
be to prepare a parallel coding
version for experienced and
advanced learners.
Our course material is mostly
visual. Any participant who
does not speak English as
their first language should be
able to follow the content, as
long as instructions are clear
and thorough. The final task
contains text but the rules are
language independent.
n Final solved puzzle

helloworld.cc 79
 LESSON PLAN

AGE RANGE
7–10 years MAKE ME HAPPY:
MACHINE LEARNING
YEAR GROUP
UK Years 4–6

LESSON TYPE

FOR KIDS
Visual / block-
based coding

REQUIREMENTS
Each pair of
Dale Lane shares a machine learning resource for primary-aged students
learners requires: n my website classroom—not just the ones who even at primary level — are able
• Project O machinelearningforkids. enjoy coding. We need to give to understand the basic principles
worksheet co.uk, I’ve built a tool that every child enough basic literacy and can build fun projects.
• Access to introduces machine learning by to be able to appreciate how the In this activity, students create a
providing hands-on experience world around them works. character in Scratch that smiles if
machinelearning
for children. It provides an easy- If they are presented in the right you say nice things to it and cries
forkids.co.uk
to-use, guided environment for way, I’m convinced that kids — if you say mean things to it.
training machine learning models
to recognise text, numbers,
KIDS ARE ABLE TO UNDERSTAND
images, or sounds.
I believe it’s important that we
“ THE BASIC PRINCIPLES OF ML
introduce this to every kid in the

STARTER:
CREATE A
CHARACTER
15 MINUTES

Firstly, students need to get

set up on machinelearning
forkids.co.uk. They can use
the ‘try it now’ function for
free — this lets students
use the site for up to four
hours without registering,
which is long enough for
most lessons.

HEALTH AND SAFETY REQUIREMENTS

Younger students may get carried away when writing
insults to train the machine learning model. It may be
helpful to set boundaries for what language is appropriate.
n Click the ‘Open in Scrach’ button. You should see new blocks from your project at the bottom of the list.

80 helloworld.cc
ACTIVITY 1: A FACE THAT REACTS 20 MINUTES n Students apply a rules-based approach to
make their character react to text.

Once set up on Scratch within the

machinelearningforkids.co.uk environment,
students will need to create faces to
respond to statements. Time management
is important here: students often lose track
of time drawing their face and don’t leave
enough time for training or coding.
Students will need to create a new sprite
with three costumes — faces that reflect
‘sad’, ‘not sure’ and ‘happy’.

n Students can type in a message

and watch the character react.

LEARNINGOBJECTIVES

Sentiment analysis

Supervised learning

n Students can use the green flag to test their

program.Type in a message and watch it react!

DIFFERENTIATION
n Write a reply: Instead of just changing the way they look, make your character reply, based
on what it recognises in the message.
n Try a different character: Instead of a person’s face, why not try something different, like an
DALE LANE
animal? It could react in different ways, instead of smiling. For example, you could make a
Dale is a Developer for IBM with a background
dog that wags their tail if you say something kind to it.
in artificial intelligence and machine learning.
He has worked on solutions for IBM clients n Different emotions: Instead of kind and mean, could you train the character to recognise
using machine learning, as well as being a other types of message?
developer for many years on IBM’s AI platform, n Real world sentiment analysis: Can you think of examples where it’s useful to be able to
Watson (@dalelane). train a computer to recognise the emotion in writing?

helloworld.cc 81
 ACTIVITY 2: SUPERVISED LEARNING 20 MINUTES

So far, students have created a character

that should react to what people type
and programmed it using a simple rules-
based approach.
However, if you want the character
to react to other messages, you need to
add more ‘if’ blocks. The problem with
this is that you need to predict exactly
what messages the character will receive.
Making a list of every possible message
would take forever!
Next, we try a better approach —
teaching the computer to recognise
messages for itself. Students will train a
computer to recognise text as being kind

SOME TIPS FOR TRAINING A MODEL

or mean. Instead of trying to write rules
to do this, they will collect examples.
These examples are being used to train a
machine learning model.
More examples
This is called supervised learning. The The more examples you give it, the better the computer should become at recognising
computer will learn from patterns in the whether a message is kind or mean.
examples you’ve given it, such as the Try and be even
choice of words and the way sentences
Try and come up with roughly the same number of examples for kind and mean.
are structured. These will provide the
If you have a lot of examples for one type, and not the other, the computer might learn one
basis for the computer to recognise and
type is more likely, which will affect the way that it learns to recognise messages.
identify new messages.
The examples your students give Mix things up with your examples
will be used to train a machine learning Try to come up with lots of different types of examples, such as, make sure that you
model. Back in Scratch, they will then be include some long examples and some very short ones.
able to use a new custom ‘recognise text
... (label)’ block which will return either
‘kind things’ or ‘mean things’ based on
the training you’ve given to the computer.
They can use this to choose the
costume to switch to. Students can then
test their projects: type a kind message
and press enter. The character should
smile. Click the green flag again. Type
a mean message and press enter. The
character should look sad. This should
work for messages that you didn’t include
in your training.

RELEVANT LINKS
Full student worksheets and teacher information sheets are available for free at
machinelearningforkids.co.uk

82 helloworld.cc
AGE RANGE
11–13 years
ASKING AI: WHAT
IS A BUTTERFLY?
YEAR GROUP
UK Year 7, US
Grade 8

LESSON TYPE
How can we introduce the concepts of neural networks to 11-year-olds?
Unplugged (pen
and paper) A resource developed for the Royal Society encourages the study of AI
Project-based
learning
achine learning is based whether images of butterflies
M on a rich tradition drawn meet the requirements of the
REQUIREMENTS
from taxonomy of species. How model. They will investigate how OBJECTIVES
• Access to the can the classification of butterflies using multiple examples builds
internet and birds help us to understand confidence in models and extends
• Pen and paper machine learning and AI? knowledge of taxonomy. What is machine
In this activity, students will When creating a neural network, learning?
learn how to create a machine there are a number of different
learning model, train it and models that characterise neural nets
investigate their findings. which can become complicated. How does machine
Most students have daily This should not be off-putting, as learning work?
contact with AI in the form of the basis remains the same.
mobile phone assistants, smart
speakers and newsfeeds. Few
have been provided with the
STUDENTS ARE PROVIDED WITH THE
opportunity to consider how
machine learning works.
“ OPPORTUNITY TO CONSIDER HOW
In the lesson, students will
isolate the characteristics of a MACHINE LEARNING WORKS
butterfly, use these to check

helloworld.cc 83
 LESSON PLAN

ACTIVITY 1: CLASSIFYING A BUTTERFLY 45 MINUTES

Ask your students to search ‘dead leaf be easier to code the characteristics, e.g. Feature Value
0 if not in list, 1 if in list
butterfly’ or ‘orchid mantis’ – images ‘Four wings = A, Six legs = B’. Step one of
which may confuse an automated building the neural network is complete.
classifier of butterflies. It would be useful The next step is to get students to view
here to get your students to consider how five images of butterflies. You could save
they would construct a taxonomy for a time by providing these yourself. Armed
butterfly. Suggest that ten characteristics with the images, ask students to write 0 in
are agreed on. This is a useful paired / the second column of circles if that feature
small group activity. Features that your cannot be seen. Similarly, students should
students should look for are characteristics write 1 in the circle if they can see the
such as four wings, six legs and so on. feature, e.g. if two antennae can be seen
then the value is 1, if less than four wings
Average =
are seen then the value is 0.
STUDENTS MODEL
“ THE TRAINING
The final stage in building the network
is to find the average of values. For a
student identifying seven characteristics
OF A MACHINE out of ten, there is 0.7 probability that
LEARNING what they see is a butterfly. Take this
further by finding the average of the
ALGORITHM averages. The process should be applied
to five or more butterfly images.
Time and prior learning may mean that At this stage, it is a good idea to take
not all of your students are able to arrive time to consider what has happened.
at ten characteristics. That’s fine, what is Students have modelled the training that n Students write butterfly characteristics in the feature circles, and
important here is that a set of features a machine learning algorithm undertakes. then mark a 1 or 0 in the second column depending on whether these
can be seen in an image. The average of the numbers in the second
is agreed. There are a range of neural net models column then gives a probability of whether the image is a butterfly.

Using the ‘Butterfly Identification used in machine learning. The maths

Machine’ students should write the behind them gets complex quite quickly confidence value. In our example, each
characteristics in the feature circles. It may but all have their basis in arriving at a image produces a value that exists
somewhere between 0 and 1.
Now it is a useful exercise to ask students
to consider a number of questions:

1. What value would be considered as

acceptable for claiming that the image
is that of a butterfly?
2. At which value would it be agreed that
the image is not that of a butterfly?
3. How many images might we need to
classify to feel secure in our model?
4. Can agreeing on characteristics allow
us as a class to share the training model
and save time?

The last question is useful in illustrating

one of the benefits of using computers to
classify images. They are fast and able to
n Students use different images to model the training
that a machine learning algorithm undertakes. engage in monotonous tasks.

84 helloworld.cc
 ASSESSMENT
Students should be able to
provide an explanation of what
machine learning is and how a
simple neural net model works

DIFFERENTIATION
Some students may need
support in searching for
images on the internet or
calculating averages.
Prepare image files
beforehand and shorten the
ACTIVITY 2: EXTENDING THE MODEL 5 MINUTES TO AN HOUR butterfly identification model.
Create flashcards or word lists
Once the training model has been agreed, they have accomplished and learned. for students to refer to.
students could repeat the butterfly Understanding the algorithm process is
identification process using images of birds. important. Time allowing, the butterfly

RELEVANT LINKS
Ask: could any of the birds be classified as a model could be recreated using flowchart
butterfly? This leads students to reconsider symbols. This could be further extended
how robust their model is and to a deeper by creating a programmed version of the
Lesson resources from the Royal
understanding of taxonomy. classification process. Society can be found at
Do we need to reconsider which In working through this topic students get
helloworld.cc/royal.
are valid characteristics? This starts to to the heart of what machine learning is. A
There is more information
introduce the concept of a multi-layer good deal of how machines learn extends on the Royal Society
neural network where the presence, or from asking the correct questions. The next resources on page 28.
absence, of key characteristics immediately step is to consider what actions need to be
renders the image as not a butterfly. For taken once the model has been trained. For
example, a score of 0 for four wings, two example, once a self-driving car is capable
antennae, thorax and abdomen is rejected of recognising a pedestrian with an agreed MICHAEL JONES
as a butterfly. Likewise a score above an level of confidence the more important part Michael is Director of Computer
agreed value for these items passes the of the process is what does the car do when Science at Northfleet Technology
image classification to the next level of faced with the pedestrian. Simply knowing College. He is a Specialist Leader
the network. that something is the case does not dictate of Education, Raspberry Pi Certified
To embed the learning of this exercise, action. This area is rich with philosophical Educator, and CS Champion for the
encourage your students to explain what questions to discuss with your students. NCCE (@MikeJonesCSTalk).

helloworld.cc 85
 LESSON PLAN

AGE RANGE
11–16 years

LESSON TYPE
AI AND ETHICS
Marc Scott shares some thought experiments your
Unplugged (pen students can discuss to explore the moral dilemmas of AI
and paper)
n this lesson, students fun, but it was really interesting to
REQUIREMENTS I will explore and discuss see the students’ thinking about
• Printed ‘Trolley some of the ethical issues that ethical problems in computer
problem’ cards are considered by programmers science, often defending some very
as they develop more advanced unusual positions.
artificially intelligent software. This is an unplugged lesson. It is
I first taught this lesson to a a great way to give your students a
group of 12- and 13-year-old break from their usual computer
students, and not only were the based lessons and get involved in a
discussions varied, energetic and lively discussion.

ACTIVITY 1: PAPER CLIPS 10 MINUTES

Start by sharing the hypothetical thought create as many paper clips as possible with the

OBJECTIVES
experiment of the paper clip maximiser with resources available. The machine then sets to
your students. First devised by the philosopher work towards a goal of turning all matter in the
Nick Bostrom in 2003, the story illustrates universe, including human beings, into either
the need to include ethics into artificial paper clips or machines that make paper clips.
Evaluate ethical
intelligence design. You can ask your students for additional rules
The story imagines you give a superhuman you should give the AI, other than making as dilemmas that AI can
artificial intelligence system the simple task to many paper clips as possible. pose to consumers and
programmers

MARC SCOTT
Marc was a teacher of Science,
Computer Science and Systems
and Control for 15 years, before
joining the Raspberry Pi Foundation,
where he now works as a Senior
Learning Manager.
 ne application of AI where ethics is debated is self-driving cars
nO

86 helloworld.cc
ACTIVITY 2: LAWS OF AI 30 MINUTES

Hopefully, in the starter activity, your as long as such protection does not You could introduce this by giving the
students will have figured out that the conflict with the First or Second Laws. students each a card with the problem, ask
single instruction, ‘create as many paper You can discuss whether these are them to read it silently and come to their
clips as possible with the resources sufficient rules or whether there might be decision before discussing it in groups.
available’, was the source of a problem. scenarios that might still cause problems. These cards could contain variations of the
The machine also needed to value human After discussing these rules, you can problem: What if there were 1000 people
life and have limits set on its resources and introduce the trolley problem: a runaway on the track? What if the one person was a
time for making paper clips. train is careering towards five unsuspecting child and the five were terminally ill? What
Now ask your students to come up workers. Do you pull a switch to divert the if instead of a lever you have the option of
with rules that all AI machines should trolley onto another track, where only one pushing someone in front of the train to
have programmed into them to ensure man works alone, or do you do nothing? stop it running into five people?
catastrophic events do not happen. How This thought experiment was devised Since its inception, the trolley problem
can they keep humankind safe from future by philosopher Philippa Foot in 1967. It has been widely used in psychological
AI machines? Ask students to share and pits two schools of thought against each experiments about ethics and morals. In
discuss these rules. other: that the morally right thing to do 2007, a study showed that 89 percent
You can introduce the three fictional laws is to save as many lives as possible so of people would push the lever, but only
of robotics, invented by the science fiction you should pull the lever, or the morally about 11 percent would push someone
writer Isaac Asimov in 1942: right thing is to never kill a person, so you in front of the train. Do your students’
• First Law: shouldn’t touch the lever. choices follow a similar pattern?
A robot may not injure a human being or,
through inaction, allow a human being to
come to harm.
• Second Law:
A robot must obey the orders given it by
human beings except where such orders
would conflict with the First Law.
• Third Law:
A robot must protect its own existence

ACTIVITY 3: SELF-DRIVING CARS 10 MINUTES

Explain to your students that there are now an increasingly large number of self-driving cars being produced and running on
our roads. Have them imagine a scenario where a self-driving car is driving at 70mph down a road, when suddenly a person steps into
the car’s path. Then present some or all of the following questions for them to answer.

Should the car hit the pedestrian or should If they were to program the car’s AI, What happens if there are three
it swerve into the wall alongside the road, whose life is more important, the passengers in the car, along with
where it might kill the passenger? pedestrian or the driver? the driver?

Should the driver be able to see the Who should decide how the cars behave? Would they be happy to be the
code the car is running and alter the Is it the responsibility of governments, the owner of a car that has been
code to make different decisions? manufacturers, the programmers or the buyers? programmed in such a way?

Recent research has shown that most people agree that self-driving cars should be programmed to save as many lives as possible.
However, they would not buy a car that they knew was programmed to sacrifice its driver. This is a problem that car manufacturers are
currently grappling with: the Mercedes self-driving car due for release this year will be programmed to prioritise its passengers.

helloworld.cc 87
CONVERSATION INSIDER’S GUIDE

ASSESSMENT IN
SECONDARY COMPUTING
Alan O’Donohoe and John Woollard offer practical, common-
sense advice on how to assess Computing for ages 11 to 16

t’s important for teachers to be clear from the start about measurements are viewed with the beginning of the time period in
I their reasons for assessment, so they can be certain that mind, as well as the end.
the assessment strategies they use are fit for purpose. Although
the motivation for assessment can be intrinsic, it’s more often n Assessment for the learner: The purpose of this is fundamentally
driven by a strong outsider influence, such as the expectations of to support each individual learner. When an individual learner
school leaders and parents or the need to meet performance can objectively observe their own progress, this will help them to
targets. We’ll look at three purposes for assessment in this article. understand where they are in terms of their own individual learning
  journey. Through regular analysis, learners can more accurately
n Assessment of learning: The purpose is to assess what the learners identify any gaps in their knowledge and understanding and take
can now do, what they now know, their improved understanding appropriate action to address these gaps. This can lead to much
and, importantly, their attitude to learning, with the intention more engaged learners in class, who are better informed about their
of recording, tracking, and reporting this to others. This type of own abilities and progress.
assessment may not be highly valued either by the teacher or the
learner, but it is necessarily fed into the accountability agenda in n Assessment for teaching: This is intended to inform the teacher in
schools, from school to school, across the school, across the subject, order to help shape their teaching plans. Assessments focus on the
and across the teaching team. These assessment results are success criteria for teaching and might include measures of learner
typically used to inform pupils and parents about the progress that engagement, attainment, and progress. However, they may also
has been made over a set time period. Therefore, it is vital that the be based on a professional judgement of what constitutes a ‘good
lesson’ and the reasons for that judgement.

It is likely that some assessment is taking place to fulfil each of the

reasons listed; it could be your own professional judgement about
how well the learners are doing when you teach in a particular way,
informal monitoring of outputs, or casual feedback to students about
their attainment in particular tasks.

Three dangers to watch out for

Danger 1: Don’t seek the holy grail of assessment. Be wary of the
dangers of being oversold apparently attractive assessment
solutions by enthusiastic advocates. Some solutions that work
extremely well, perhaps for a teacher in one particular context with
boundless energy and motivation, are probably unsustainable in the
long term. In reality these solutions may have:
n High start-up costs in terms of investing time to learn to use the

system and adapt the resources

n Regular assessment can lead to more n High ongoing maintenance costs in capturing and inputting data
engaged learners
n Low value in terms of informing stakeholders

88 helloworld.cc
n Don’t forget to consider the impact of your chosen solution on your workload

“ DON’T DISMISS SIMPLE

SOLUTIONS, AS THESE MAY
BE THE ONES THAT WORK
BEST FOR YOU

It’s really about finding an approach that you think will work
for you and your learners in your setting, and then refining and
developing it to get the most from it. If this is the way you approach n We have suggested four tried and tested assessment strategies

assessment, then the holy grail will eventually come and find you!

Danger 2: Consider the workload impact. Your well-being is incredibly ALAN O’DONOHOE
important, and there is little value in creating a wonderful-sounding Alan has more than 20 years’ experience teaching
assessment strategy if it is impractical to use on a regular basis. and leading technology, ICT, and computing in
It would be wise to calculate how much time would be needed schools in England. He runs
for you to keep on top of a solution, and then consider this in exa.foundation, delivering professional
terms of all the classes you teach. ‘Little and often’ is the key with development to engage digital makers, supporting the teaching of
assessment. Do not dismiss simple solutions, as these are often computing, and promoting the appropriate use of technology
the ones that will work best for you. Solutions that are ‘just good (@teknoteacher).
enough’ are the most likely to serve you well in the longer term. You
can always tweak and improve a system that is just good enough JOHN WOOLLARD
once it becomes embedded into your regular routine. John Woollard is a teacher trainer, author, and
university lecturer working in both computer science
Danger 3: Beware false success. Some commercially available and special educational needs. He is a member of the
assessment solutions do not limit or control students’ attempts BCS School Curriculum and Assessment Committee,
at multiple choice quizzes, meaning that students can choose to and chair of CAS Assessment working group.
repeat the tests until they score 100%. These can be popular with

helloworld.cc 89
CONVERSATION INSIDER’S GUIDE

n It’s all about finding an approach that will work for you and your learners

students, who love the rewards of achieving high scores in tests.

However, this creates a dangerous, false sense of success in which
students have only learnt the answers to the tests by rote, and this
learning does not convert to the same success in an exam that is
“ MULTIPLE CHOICE QUIZZES
SHOULD BE A PROMINENT
designed to assess their understanding. If you are using one of
these systems, take steps to prevent your students from repeating FEATURE OF ASSESSMENT IN
the quizzes too many times. It’s best not to use a particular question
or quiz more than three times a year.
ALL COMPUTING CLASSROOMS
Recommended strategies Attitude assessments
In this article, we are recommending four strategies to consider. Every week, students are prescribed a set topic to study out of
class. This is described in depth in the ‘No Headaches Homework’
Multiple choice quizzes (MCQs): Instant quantitative results article in Hello World issue 8. Students are expected to provide
Multiple choice quizzes should be a prominent feature of evidence of their studies in an agreed format every week, and this is
assessment in all Computing classrooms. Computer-marked graded out of five. If the teacher collects the books for grading every
assessments provide teachers and learners with a rapid means of two weeks, it’s possible to assess both pieces of work from a single
achieving instant quantitative results that can be used to inform class in less than 30 minutes. Over time, this generates another set
learners of their progress and provide data to help teachers of numeric data that teachers can use to inform their assessment
diagnose gaps in understanding. It’s important that the same judgements. As well as providing a rich programme of learning out
questions are used before a period of study as well as after. There of class, this strategy provides teachers with some valuable data
is an initial time investment in importing or creating the quizzes, but about students’ attitudes to learning.
once they are set up, the software automates the assessment. If
these are used regularly, over time they will help paint a picture of Markbook
the learning landscape in your classroom. This is a subjective assessment that has the benefit of providing
long-term data. At a convenient point during each lesson, take two
Exam-style assessments minutes to scan the class and put a mark against each pupil’s name: 
These are short, one-page exams lasting just ten minutes, n 0 means absent

presented in the style of an exam paper every four lessons, with n 1 means present but with little effort or achievement

a variety of questions worth one, two, four, and nine marks, to be n + means meeting expectations (whether it be for the pupil,

completed either in exercise books or on paper. After the time has class, or year group)
elapsed, students swap papers and mark them in class under the n * means exceeding expectations

direction and supervision of the teacher, who then collects the Handwritten entries are easy to change: in this system, a ‘1’
papers for moderation. These give the students a more objective becomes a ‘+’ by adding an extra line; a ‘+’ becomes a ‘*’ by adding
form of assessment in an exam-style experience; while these are an extra two lines.
not as easy to administer or assess as MCQs, they do provide In this strategy, we make professional judgements based on what
valuable data that feeds into the assessment of learning. we see, but importantly, these judgements are tempered with what

90 helloworld.cc
we know about the individual learners, the task being undertaken,
and the context of the lesson. This is a classic ‘assessment without
levels’ approach. All judgements are made relative to that pupil, in
that class, at that point in time. 
The record reveals trends over time, both for the class and
n Aim to make the assessments you choose a habit, so
individuals, reveals topics that are not doing well, and, importantly, that you rarely need to give them any further thought
provides a good aide-memoire or baseline when the next stage of
reporting takes place.
Assessment records in the form of spreadsheets or forms can be
used for professional-looking communication with the subject lead
or senior management.

Make assessment a habit

We have suggested four tried and tested strategies that you could
use in your own teaching. Each of them can work very well in
isolation, so you could start by introducing one of them into your
teaching and concentrate on making it work for you before you
begin trialling any of the others.
Whatever approaches you choose to use for assessment, aim to
arrive at a position where the assessments become a habit, so that
you rarely need to...
n Give the processes further thought; 

n Provide additional time and/or effort; or

n Reintroduce the system to the learners

For more free, friendly advice about assessing Computing, contact

the authors of this guide, Alan O’Donohoe, [email protected],
and John Woollard, [email protected]. n A markbook is a classic ‘assessment without levels’ approach

helloworld.cc 91
 CONVERSATION QUESTIONS

YOUR
QUESTIONS
Q DO YOU HAVE ANY
SUGGESTIONS FOR
UNPLUGGED LESSONS
TO USE WITH A YEAR 8
CLASS WHO ARE ABOUT
TO START PROGRAMMING?
BEN GARSIDE: Getting learners to understand that
A computers need precise instructions is a key objective
before they start programming. There are some really As a shorter unplugged activity later in the unit, you might
interesting offline activities you can do around this, such consider using Parsons problems (where learners have to
as getting learners to direct a human robot around the reorder jumbled up lines of code to form a working segment
classroom using instructions, whilst you watch out with of code). They are a great way to recap the learning from a
a keen eye for the learners making human assumptions. previous lesson and could be done independently or in small
Another activity that is a favourite of ours at Raspberry groups, before going to the computer and testing out their
Pi is for learners to form a 'human algorithm' where they solution (see page 87 of Hello World issue 8).
have to perform sounds in a sequence. Both of these I’d also recommend checking out the following two
activities feature in lesson 1 of our Year 7 programming websites that have some great ideas for unplugged
unit on the resource repository (helloworld.cc/evidence) lessons: cs4fn.org ('Searching To Speak' is a personal
and are described in detail. favourite of mine) and csunplugged.org.

Q WHAT ARE YOUR VIEWS ON COMPUTING EDUCATION AND

SCREEN TIME FOR STUDENTS IN PRIMARY SCHOOLS?
SWAY GRANTHAM: Both adults and children spend a n content (what is being watched or used)
A lot of time using screens in their daily lives. Most of the n connections (whether and how relationships are
recommendations around a limited time using devices, facilitated or impeded)
such as the World Health Organisation’s advice, is In the modern world, limiting screen time could mean
actually around the idea that screen time is sedentary limiting the time someone has to talk to a family member
time, which other activities, such as reading, also involve. or practising their times tables! It’s really important that
However, research such as the LSE Media Policy Project, we all participate in a variety of activities, both online
which reviewed current advice and upcoming research and offline. In our curriculum resources for year 2, we
on screen time and related topics, shows the concept is introduce children to the Children’s Commissioner's
totally misleading. Digital 5 a day which suggests five activities you should
They suggest instead that it’s about: do each day, to ensure you have a healthy balance and
n screen context (where, when and how digital media are not spending too much time doing one of those
are accessed) activities, whatever that activity is.

92 helloworld.cc
 Q I’VE GOT A SPARE BLOCK
OF TIME WITH MY YEAR
8 CLASS AFTER THEY
LEARN PYTHON. WHAT
SHOULD I TEACH?
BEN GARSIDE: If you’re interested in app design,
A I recommend that you take a look at App Lab
from code.org. It is a free-to-use, web-based
coding environment where learners can use
drag and drop tools to create the user interface
elements to the app, but use code to power the
app. The code interface is block based by default
but can be switched to text based where the
learners can code in JavaScript. Code.org have
an 'hour of code' activity to allow learners to
become familiar with the interface, but keep an
eye on the NCCE Resource Repository where
we will soon release a full unit of work on app
design using App Lab.
MARTIN O'HANLON: If you wanted to take
their Python knowledge further, you could
consider using Python to create graphical
user interfaces (GUIs) or introduce physical
computing. There are two online courses
available via teachcomputing.org, Programming
If you have a question you’d like the Learning Team at the
Raspberry Pi Foundation to answer contact us on Twitter via with GUIs and Teaching Physical Computing
@HelloWorld_Edu. Alternatively, email us with ‘Question’ in with Raspberry Pi and Python, which will give
the subject line at [email protected]. you ideas and materials to get started.

Q MY SCHOOL IS KEEN TO FIT IN COMPUTING INTO WHOLE-DAY

SESSIONS. DO YOU HAVE ANY RECOMMENDATIONS ABOUT
WHETHER COMPUTING SHOULD BE TAUGHT IN DAY-LONG
SESSIONS OR BLOCKED OVER A WEEK?
SWAY GRANTHAM: There are a couple of things to consider weekly lesson, a teacher would reflect on their teaching and
A when blocking the curriculum. Firstly, are you confident that what the children learned and make amends to next week’s
the curriculum coverage will be the same. For example, if it’s plans accordingly. However, when the next lesson is in 10
usually a six week half term and you’ve only got five hours minutes, it’s unlikely the teacher has the time or resources
in your computing day that half term, how will teachers to change the plan for the lesson if something isn’t going
manage that time limit whilst ensuring there are no parts of well. I also wonder, if the learners will find a day of building
the curriculum left out and all are taught sufficiently? on the same concepts challenging for their cognitive load,
Secondly, by teaching something in one solid day you which may mean they don’t retain what they learned as
have fewer Assessment for Learning opportunities. In a successfully. However, I have not tested this theory.

helloworld.cc 93
 CONVERSATION LETTERS

YOUR Our letters page is a place for you to join our conversation.

LETTERS
If you’ve got a comment, a question, or an announcement
to share, contact us on Twitter via @HelloWorld_Edu.
Alternatively, email us with ‘Letter’ in the subject line
([email protected]).

A feeling of home
Hello, HelloWorld!
I'd like to just send a quick email to say, "Thank you!"
In the world of Computer Science education, there aren't
many avenues for connecting with peers, especially in the
U.S., where it's not a mandatory subject area. Often, you're
the only CS teacher/advocate at your school, so opportunities
to connect are very limited. The internet can be a big and
overwhelming place to find comradery and resources that are
a perfect fit. Online searches can steer you in directions far
from what you're truly searching for. With all of this, it's so
easy to feel like you're all alone, with sporadic places to turn
to piece together resources, best practices, model lessons and
advice from peers.
When I read HelloWorld, I get a sense of feeling 'at home';
I've found my people. There's something in each issue for
everyone related to CS education. From teachers to coaches,
to policymakers, to advocates, to researchers, and more;
everyone can get something from each issue. In my role, I
support CS teachers through professional development and I
find myself constantly using articles to encourage, challenge
and motivate my participants.
Thank you again for all the efforts that go into bringing this This email made our day, thank you, Rosemary! You’ve summed
to life. It's truly an invaluable resource for continuing to grow. up exactly what we try to do here at Hello World.
Rosemary Bianchi Sian Williams Page, Editor

HELLO WORLD WRAPPING: FROM PLASTIC TO PAPER

I’m really happy to say that from this issue onwards, Hello World will be will account for 20 per cent of the world’s total oil consumption,
shipped to our hardcopy subscribers in paper envelopes. exacerbating the climate crisis.
As I wrote in response to Paul Sutton’s letter in issue 9 (page 95), Some readers suggested we use decompostible potato starch
the move away from plastic packaging has been something we’ve been wrapping to combat this, used by several other publications. However,
trying to implement for a while now. not everybody has access to a compost bin. Paper recycling is
A recent UN report highlighted that plastic packaging accounts widespread: hopefully it’s clear to know what to do with this envelope.
for about half of the plastic waste in the world. When discarded It can often feel like individual actions are futile in battling the
in landfills or in the environment, this can take up to a thousand environmental crisis but it’s thanks to the readers’ letters that we were
years to decompose, causing numerous problems as it does so. High able to justify this switch in packaging. I know I will bear this in mind
concentrations of plastic bags have been found blocking the airways and when considering how I interact with other businesses and brands in the
stomachs of hundreds of animal species. The toxic chemicals in plastic future. Thank you to everyone who has emailed, or left a comment on our
packaging transfer to animal tissue, eventually entering the human food survey, to ask for this change.
chain. And, if current patterns continue, by 2050 the plastics industry - Sian Williams Page, Editor

94 helloworld.cc
OUR
SURVEY
RESPONSES FROM OUR 2019 READER SURVEY
Thank you to everyone who took the time These will only be sent to a subset of our
to answer our reader survey. The results subscribers — the research team at the
are really important to us: we are constantly Raspberry Pi Foundation assure us that
trying to make Hello World as useful as random sampling leads to more robust
possible for computing educators and your results — so if you receive an email from us
feedback guides us on how to do this. asking for feedback, please do take a few
This year, we are trialling using shorter minutes to tell us how we can improve the
surveys at more frequent intervals. magazine for you!

I find Hello World is

sometimes too focused on
one subject rather than a
mix of topics. Some of the
editions I haven't really
read much because of this,
particularly those with long
articles which I don't feel
are that relevant. I prefer it
when there is something for
everyone. I do enjoy learning
about what teachers are
doing in other schools but
prefer the shorter articles.

Thanks for this feedback, we

try to strike the right balance
between features that can
delve into the details of an
idea and easier quick-reads.
It’s something we’ll carry on
thinking about.

Excellent magazine to learn what is happening in the computing world for

educational purposes. I may not get lessons directly from the magazine, but it helps
to generate other ideas that I may not have got without reading the magazine. I find the magazine a
I appreciate that the magazine is less frequently published and has more great source of ideas and
interesting content. It's not just trying to sell something. I wait for the publication resources that complement
but I take a good month to work through it. my teaching. Could you have
I don’t often get the time to read the magazine. A digital version would be great a SEN bit about teaching
as a PDF seems not very accessible: I can’t easily start reading one or two articles. I kids with different abilities?
guess I am more of a snacking person when it comes to reading stuff online, rather
than a magazine person. I’m really happy to say
Catherine Elliott has started
We are starting to share articles online from this issue onwards. Take a look at her regular column on SEN.
helloworld.cc.and let us know what you think. See page 14 of this issue.

helloworld.cc 95
 I find the magazine a great source of ideas and
resources that complement my teaching.
I have found it very useful as a catalyst to investigate
new things or further my knowledge on familiar things.
I live in the US and would like a reasonable cost
option to subscribe to a print version.

Great to hear you enjoy the magazine. A print version in

the US isn’t on the cards in the immediate future, but you
never know!

I Love Hello World! Such a fantastic resource for anyone teaching CS!
And thanks for making digital editions available for those of us teaching It looks like the targeted
outside the UK. audience for this magazine
I love the magazine, and I especially like when there are articles is teaching age 6-14. I miss
showing specific projects that I can bring to the classroom, in particular what should be taught after
Scratch programs. this age group (the age of
I think it's an absolutely brilliant resource. I think the thing it does my students are 17-18 yo)
for me is remind me that I'm not alone and that it's ok to try and fail
and learn as there is a community of support. It's a rare publication, Resources for and features
something that makes me feel part of a community of practice and around teaching A Level
that's why it's important to me. I may not read all the articles but when students is something we can
I see it come through the post box it reminds me that there are good definitely make more space
educators out there doing great things and that I too can contribute to for in the magazine. In the
this in some small way meantime, if you’re looking
I think it's an excellent resource. Please keep them coming. There's for ideas for how to support
something about holding a copy of the magazine in your hands that's, these students check out
uh, intangible online. isaaccomputerscience.org

96 helloworld.cc
 REVIEWS BOOKS

BODY IMAGE IN ESSENTIAL READING

THE PRIMARY SCHOOL Three books to help you explore
machine learning further
Ben Hall discusses the second edition of a
book dealing with one of the most pressing
DATA SCIENCE FROM SCRATCH
issues for children in the modern world
BY Joel Grus
BY Nicky Hutchinson and Chris Calland | PUBLISHER Routledge | PRICE £23.99
IN FO ISBN 978-0-18842-581-9 | URL helloworld.cc/image
PUBLISHER O’Reilly
PRICE £29.85
ISBN 978-1492041139
URL helloworld.cc/datascience
W hen Nicky Hutchinson and
Chris Calland published
chapters offer a clear explanation
of body image, how it can affect Learn how many of the most
the first edition of this book children, and how it can be fundamental data science
in 2011, they could not have incorporated into the curriculum for tools and algorithms work by
foreseen how much more pressing children between 4–7 and 7–13. implementing them from first
principles. Learn the basics of
its subject matter would be in There is then a fully resourced
linear algebra, statistics, and
the nine years that followed. scheme of work, including lesson probability and understand
In 2011, the smartphone was plans with extension activities. The how and when they’re used in
in its infancy, iPads was less plans are refreshingly simple and data science.
than a year old, and Instagram easy to follow, a welcome contrast
celebrated its first birthday with to other, more wordy PSCHE
AI NARRATIVES
fewer than 10 million users. Fast- schemes I have come across. The
forward to 2020 and we have the age split of the lesson plans may be
BY Stephen Cave (editor), Kanta Dihal
latest Ofcom statistics showing an issues for some teachers: 7 to (editor), Sarah Dillon (editor)
that 45 per cent of 5–15 year 13 is a broad range, and in the UK PUBLISHER
olds own their own smartphone, this splits across two Key Stages. Oxford University Press
and double that number have PRICE £55
ISBN 0198846665
online access at home. With A sensitive approach URL helloworld.cc/narratives
these devices enabling unfiltered Central to the approach included
access to images of ‘perfect’ in this scheme of work is an The potential social, ethical,
political, and philosophical
bodies, perpetuated by the media, avoidance of overt references
consequences of AI are
the rationale for this publication to diets, junk food, or body explored in detail, with
couldn’t be any clearer. appearance. Instead, it focuses on contributions from leading
The book itself is split into two promoting emotional well-being humanities and social
main sections — the introductory and resilience. These are exactly science scholars.

the qualities children will need to

deal with and process the stream
THE BOOK OF WHY
of artificial perfection presented
to them on their social media BY Judea Pearl and Dana Mackenzie
feeds. This is a sensitive and PUBLISHER Penguin
well-thought-through strategy — PRICE £10.99
ISBN 0141982411
there is a substantial amount of
URL helloworld.cc/why
research suggesting that tackling
such issues head-on can make the Judea Pearl’s accomplishments
problem worse. over the last 30 years have
While it may have been difficult provided the theoretical
to anticipate how much changes in basis for progress in artificial
intelligence. Here, Pearl and
technology and social habits would
science writer, Dana Mackenzie
bring body image issues to the explore the saying ‘Correlation
forefront, there is no doubt that this does not imply causation.’
problem is now here to stay.

helloworld.cc 97
 “HELLO, WORLD!” Everything you need to know about our computing and
digital making magazine for educators

Q WHAT IS HELLO WORLD? Q WHO MAKES

Hello World is a magazine for computing and
HELLO WORLD?
A digital making educators. Written by educators, The magazine is a joint collaboration between
for educators, the magazine is designed as a platform to A its publisher, Raspberry Pi, and Computing at
help you find inspiration, share experiences, and learn School (part of BCS, the Chartered Institute for IT).
from each other.

Q WHY DID WE MAKE IT?

There’s growing momentum behind the idea of
A putting computing and digital making at the heart
of modern education, and we feel there’s a need to do
more to connect with and support educators, both inside
and outside the classroom.

Q WHEN IS IT AVAILABLE?
Your 100-page magazine is available four times
A per year.

IT’S FREE!
Hello World is free now and forever as a Creative
Commons PDF download. You can download every
issue from helloworld.cc. Visit the site to see if
you’re entitled to a free print edition, too.

98 helloworld.cc
 WANT
TO GET
INVOLVED?
There are numerous ways for you to get involved with the magazine.
Here are just a handful of ideas to get you started

Give us feedback Tell us your story

Help us make your magazine better – Have you had a recent success (or failure) you
your feedback is greatly appreciated. think the wider community would benefit from
hearing about? We’d like to share it.
Ask us a question
Do you have a question or a bugbear you’d Write for the magazine
like to share? We’ll feature your thoughts Do you have an interesting article idea or
and ideas. lesson plan? We’d love to hear from you.

GET IN TOUCH
Want to talk? You can reach us at:
[email protected]

FIND US ONLINE SUBSCRIBE

www.helloworld.cc IN PRINT
@HelloWorld_Edu TODAY!
PAGES 34–35
fb.com/HelloWorldEduMag

helloworld.cc 99
 helloworld.cc

99 (helloworld.cc)