Jonny Austin

Should we really be trying to make our tech indistinguishable from magic? Is there dark magic in tech? And what does that have to do with a teleporting duck in a classroom?! We often hold creating 'a magical experience' as a key goal for design - it's fun, engaging, sometimes even playful, and it makes complex things look simple. But technological magic has a dark side too, especially when trying to help people learn and feel confident about technology. By its nature, magic you experience is not under your control: it's a trick, and you're not supposed to be able to understand it; magic is inscrutable, and for many, that's disempowering. In designing the BBC micro:bit and surrounding tools, we've thought a lot about the balance between the positive and the negative sides of creating of a magical experience. This talk will reflect on our struggle to balance technical authenticity and honesty with the need to provide a high quality experience that excites and inspires students. It will explain how we attempt to present simple, understandable analogs of more complicated, magical technology in order to help students gain a sense confidence with the tech around them. However, it will also look at how we've resorted to using our own bits of (nearly) invisible magic to make this all work nicely in a classroom. Through this discussion we'll look in detail at the presentation of the micro:bit's IO, the way the micro:bit's USB interface works, the microphone privacy LED, the web-based compiler and the simple radio communication. Finally, I will look beyond micro:bit at how using this concept of balancing transparency and magic can help us build better, more trustworthy tech.

The world is being shaped by machine learning. In fact, lots of the pages you visit on the internet are curated by algorithms that have learned what you like, or might buy. But what's actually going on? And how do machines 'learn'? In this workshop we'll use the BBC micro:bit to make everyday objects that you bring along, or people that you attach a device to, smarter with machine learning. Make a tennis racket that knows whether you've played a forehand or a backhand, or a shoe that knows whether you're walking or running. Or perhaps a device that can detect your dance moves. By going through the whole process of collecting data, training a model and then testing the model, we'll demystify some of how machine learning works in the real world, and hopefully have some fun throwing micro:bits around doing it. We're testing some new software for use with the micro:bit, and this is one of our first outings with it. Please come with the spirit of adventure. Families welcome as long as there are some members who are keen to lead the others! Suitable for children under 8 if helped by an adult.