Designing a Weight-Responsive Self-Calibration Algorithm
For these reasons we had to think very carefully when
designing the algorithm which mapped the children’s
actions to audio-visual responses. We had initially
considered working with a fixed set of ‘ingredients’ that
could be combined to produce a finite number of predefined
‘spells’. This idea was rejected for reasons both practical
and pedagogical. Logistically, we acknowledged the
likelihood that items would routinely go missing, and that it
might be difficult to find exact matches for their identical
replacement. Equally, from a pedagogical perspective, we
were aware that this form of interaction could overly
constrain children’s natural creativity.