"Terence! Terence! Why doesn't this work?

"Help! My bird isn't flapping!"

"What do I press to set this variable?"

I am run off my feet. I dash from workstation to workstation, troubleshooting the problems. In every case, my response is the same.

"Did you follow the instructions on the sheet?" I say.

"YES!" Comes the howl of protestation.

"Even this one where it tells you to click on cat before making the changes?"

"Oh... Oh! That makes sense! Thanks." and off they go.

I'm quite serious. For five lessons, I extolled the virtues of reading the worksheets. Of double-checking one's workings. Of re-reading the instructions and making sure they had been followed.

No impact.

So, I bought out the big guns!

Yeah, that's right. Biro pens!

"Next to every step," I intoned, in my most schoolmasterly voice "is a check box."

I showed them the worksheets created by Code Club. "Once you have completed a step," I continued, "You are to tick the box with your pen. Then, and only then can you proceed to the next step. And woe betide any child who calls me over for help without first having made sure they have completed all the necessary steps..."

I then set them off and braced for the inevitable flurry of hands.

Nothing! Blissful silence as the kids toiled away over their Scratch projects.

I waited and waited. Slowly, a hand raised.

"Terence," the girl's voice wavered, "I've instantiated my variables but I don't know whether they should have global scope or merely be restricted to local scope."

(Ok... ok... She didn't word it quite like that - but that's what she meant.)

A few minutes later a boy stuck up his hand.

"I've found an interesting race condition which occurs when two scripts are initialised simultaneously. Is there any way to guard against this?"

(Again, I've embellished. Look, these are 9 year olds learning Scratch. Their computer science vocabulary is somewhat limited.)

In fact, there were only a couple of questions which could be answered by saying "read the text thoroughly and tell me what you've missed."

The majority were interesting questions about Scratch, how to craft fun games, general computing problems, and the philosophical nature of whether a horse could outrace a bat.

And all because I'd handed each kid a pen and taught them how to use it.

Ticking All The Boxes

By complete coincidence, I have just finished reading The Checklist Manifesto by Atul Gawande

Much like Malcolm Gladwell's books, its premise can be summed up in a single sentence. People doing complex tasks often forget basic steps - following a checklist helps prevent errors.

In Atul Gawande's book, he talks about his success at introducing a checklist at hospitals. By reminding surgical teams to make sure they were operating on the correct side of the patient, had appropriately sterilised the equipment, and knew of any likely complications - they were able to prevent thousands of mistakes and save hundreds of lives.

It sounds too good to be true, doesn't it? Having a tickbox which says "check patient has suitable anaesthesia" (or whatever) actually works. To quote from a study of the checklist published in the New England Journal of Medicine:

The rate of death was 1.5% before the checklist was introduced and declined to 0.8% afterward (P=0.003). Inpatient complications occurred in 11.0% of patients at baseline and in 7.0% after introduction of the checklist (P<0.001).

You can find links to some peer reviewed articles about the success of the checklist on Gawande's website.

While I recommend buying the book - you can read for free the original article which inspired it in the New Yorker.

It's important to realise that a checklist isn't a universal panacea. The checklist itself has to be well designed and easy to use. But after seeing how it transformed a cluster of kids in a Code Club, I'm a convert. Before, they struggled to reach the end of a worksheet because they spend so much time correcting their mistakes. With the checklist, they were easily able to reach the end and could spend more time playing, coding, designing, and having fun.

More importantly, the kids didn't seem to mind having to occasionally tick a box to move on to the next step. It didn't interrupt their flow or stifle their imagination.

An article in the British Medical Journal about the effectiveness of the checklist in surgical situations concluded with:

The checklist was considered easy to use by 80.2% of respondents, while 19.8% felt that it took a long time to complete, and 78.6% felt that the programme prevented errors.

The real kicker is the final line.

Overall, 93.4% would want the checklist used if they were undergoing operation.

If you're teaching kids or adult - consider seeing what a simple checklist can do for your students.

The first lesson was a great success. It was held before the Christmas break - so I was eager to make sure that interest hadn't waned. Far from it - we were over subscribed!

We had 10 PCs and 12 kids - which lead to some sharing of equipment. Everyone involved seemed to enjoy it (including me) - so I thought it would be worthwhile to make some general notes which I hope may be of use to other instructors.

• Girls like coding! Around a third of participants were mini-LadyGeeks. The CodeClub plans are suitable for boys and girls.


• Sharing a PC sucks. I don't know if it's better to exclude kids who turn up late, or make them do "pair programming" - but it's harder to learn if one person is doing the work and one is an onlooker.


• Make sure you have all your handouts ready. Due to a printer mishap, we had to share some worksheets - that's not great when kids are working at different speeds.


• Scatch is very simple. Which means the enterprising child can find half a dozen ways to make it misbehave! With 10 children clamouring for attention, it can be hard to spot where the bug is. Sometimes it's best to say "Delete everything, start again, and follow the instructions really carefully!"


• Some kids finish early and want to be stretched. You can either give them the advanced assignments (add more monsters, change their colours, etc) or you can deputise them into classroom assistants. Guess which option I chose!


• Kids can be really creative - both in terms of the customisations they build and the questions they ask. I'll never forget how a little girl with an Alice band asked me "How can I make the monster drip with blood and then have its head explode when I click on it?"


• Parents can be a hindrance. Remember when you got a train set for your birthday and your dad "helped" you by putting it all together, breaking it, and then claiming it wasn't his fault that you weren't enjoying it? Any kids who can use a mouse can probably get by without parental help.


• That said, having parents view the child's work at the end of the lesson is really encouraging. It lets the parent know what their kid has been doing and it gives the kid a chance to show off.


• Have a memory stick for each participant to save their work on. Thanks to Telefonica (my employer, I'm not speaking for them, may contain nuts, etc.) I was able to give a flash drive to each kid. I put on a copy of Scratch for Linux, Mac, and Windows - along with some other programming resources. Not every kid will have a computer at home, but Scratch will quite happily run from a memory stick on a school or library machine.


• Use headphones if the program needs sound. It stops the kids from getting distracted by hearing the noises eminating from their friends' PCs. In a library setting, it stops you getting glared at.


• Have helpers. I'm enormously grateful to the staff of Woking Library who have been helping me. Dealing with ten excitable children is slightly more complex than I imagined!

...and all that's just from two sessions! Wonder what the next will bring?

If you want to help young people learn how to program computers, you can download Scratch for free - and register with CodeClub to get their lesson plans.

This is the blog version of the discussion I lead.

Kids And Phones

Kids love their phones. Can we use that love to encourage them to learn how to code?

Here's a great stat about teens and mobile phone ownership:

Two-thirds (65%) of children aged 8-15 own a mobile phone

• 49% of 8-11s
• 82% of 12-15s.

Pretty stunning stuff, I'm sure you'll agree. The only problem is, those statistics are from 2006!

Ofcom has been commissioning studies into phone ownership for several years. They give us a great insight into phone usage among kids.

For example, more recently, we see this gem of a stat.

The correlation between age and mobile phone use is particularly strong, with the proportion of children using a mobile almost doubling between the age of 9 (52%) and 15 (95%).

[...] children are acquiring mobiles at a younger age and using them more.

Again, wow! 95% of kids have a mobile. Oh, that stat is from 2008!

Let's go bang up to date with the most recent Ofcom study.

Smartphone ownership in 2010 comprised

• 3% of 5-7s,
• 13% of 8-11s,
• 35% of 12-15s.

A third of kids have smartphones. I'm not sure that we need a whole bunch more statistics to tell us that mobile phones - especially smartphones - are highly desirable to kids. Both boys and girls.

Click to Embiggen.

Social Exclusion

PC ownership isn't as rare as it once was - but it's still a pretty big barrier to entry. Especially if you have to fight parents and siblings for time on a shared computer.

Phone ownership is fairly wide across all levels of society. As this report shows. Click to embiggen.

Which Phone?

Before running off into discussions about what should be taught - let's take a look at which platform is the best for kids.

Taking into account kids' smartphone ownership rates, we get a chart like this. While BlackBerry is very popular at the moment, Android has already started to eclipse it.

The rise of Android is unstoppable when you consider how cheap the phones are getting.

This is the Huawei Ideos:

The Ideos currently retails for US $80 in Kenya. Eighty bucks for an Android 2.2 smartphone. With 3G, Wifi, GPS, touchscreen, and all the other fun stuff the more expensive phones have.

Here in the UK, we're seeing more handsets come in at the sub £100 mark. There are going to be a lot of these cheap but capable smartphones in Christmas Stockings this year.

However, even at £100, that's still too expensive for many students. But I don't think it's too expensive for schools.

Back in the mists of time, when I was a child, our family was lucky enough to own a BBC Micro. This was one of the first mass-produced computers intended for educational user. Most schools had at least one kicking around. But they weren't cheap.

In 1981, the BBC Micro Model A cost £235. The B Model cost £335. (Source: Wikipedia's BBC Micro Page

Adjusted for inflation, that's £750 and £1,050 respectively. A huge quantity of money.

At £80 per phone, a school could quite easily buy a couple of Android handsets to go in each computer lab - and, perhaps, let them be loaned out to students.

Android Advantage

There are two other app development platforms worth considering; BlackBerry and iPhone.

While BlackBerry is popular (for now) it is an absolute pig to program for. It has a level of complexity that is just painful. Added to the problem is its ridiculous signing requirements and you end up with a platform with is quite unusable for kids.

iPhone is the darling of the industry - and very easy to code for - but has three very serious drawbacks.

1. Expensive. iPhone ownership is low because the hardware costs are so high.
2. Limited platform. You can only create iPhone apps on a Mac. If your school has a few hundred Windows XP computers - you'd have to replace them all with expensive Macs.
3. Hard to distribute. If you've written an iPhone app, it is relatively hard to put it on a dozen iPhones - unless you want to get it certified by Apple, which could take a while!

So, Android has the upper hand.

1. The development kits is available on Windows, Mac, and Linux, and will run on very modest computers.
2. Android programming is done in Java. Lots of free tutorials out there.
3. 3D programming is available with OpenGL.
4. There are no distribution restrictions. If a kid has written a fart app - they can spread it round the school to their heart's content, using nothing but BlueTooth if they wanted.
5. Opportunity to make money. If you want to distribute or sell your app through the Android Marketplace, the registration fee is only $25. A school or club could easily register and get all their kids work up on the global stage.

So, how do we capitalise on that and get them interested in coding?

I think that, rather than getting kids to program big boxy computers - where they can only make use of their creations in the computer lab, or a PC they share at home - we should get them writing apps.

Not Angry Birds (although that's great from a physics & mechanics perspective), and not FourSquare clones (although, again, great for geography course work), and not InstaGram filters (could be useful in an art class, I guess), and certainly nothing like Rosetta Stone (may be useful if they're learning a foreign language, perhaps)...

Ahh.... you see, apps can be useful in most classes. Not just for an optional "programming" module.

More than that, they're cool. I don't know if kids still say "cool" - but apps have a high social cachet. "What's that app you're using?" "Oh, just something I wrote!".

Yes, we can start off on "Hello World", then fairly trivially move on to inputs, then conditional statements, then reading sensors, then drawing graphics...

Before you know it, kids have built an app which plays a Justin Beiber clip every time they answer a question correctly in a history test - but only while they're in the playground.

Isn't Someone Already Doing This?

Indeed they are! The CDI AppsForGood team go in to inner city schools and get kids to design apps which will be useful to them and their peers.

I was rather dismissive of the idea when I first heard about it - but I have since seen the error of my ways.

So, my CodingForKids pledge is two-fold

1. Work with AppsForGood to help them achieve their aims. Be an expert advisor to their students. Involve the companies I work for with the project.
2. Find and promote Android programming resources specifically designed for kids. Promote apps which have been built by young students.

We Need You!

If you want to get involved, please visit the CodingForKids Wiki and see how you can help.