Most shelving relies on rectangular compartments that ignore the contours of the objects they hold. We present Uniquely Shaped Spaces, an object-driven algorithmic tool for custom shelving generation. The workflow arranges users’ object silhouettes with simulated annealing, grows walls via cellular automata to carve fitted voids, and outputs fabrication files with joinery for laser cutting. We designed the system so that objects, our algorithm, and users share authorship, and studied how this configuration played out with five participants as they designed shelves in guided workshops and then lived with the fabricated pieces. Our findings show how participants navigated object geometry, algorithmic search, and fabrication limits by curating, tweaking, and appropriating algorithmic proposals, and how the resulting shelves supported reflection and storytelling. These results point toward object-driven fabrication systems that foreground objects as generative constraints and explicitly support negotiation within constraint-driven workflows.

Clay 3D printing is a relatively new technology and only a narrow range of geometries is 3D printable if one is employing commercially available slicing software. We experienced these limitations in an artist residency program where artists discovered that many desired geometries failed to print successfully. This motivated us to develop WeaveSlicer, a slicer optimized for 3D printing in clay that maintains constant wall thickness throughout the form. We achieve constant wall thickness by generating an oscillating path where the amplitude of the oscillation is determined by the form’s overhang angle. We demonstrate the effectiveness of our approach by comparing a range of successful prints, sliced by WeaveSlicer, to failed prints of the same forms sliced by Cura, a widely used slicing software. We then showcase a collection of complex artifacts designed by artists in residence that were constructed with WeaveSlicer.

We present our experience with the design and implementation of BJC Sparks, a new introductory computer science curriculum for middle school piloted by ten teachers in the 2021-2022 academic year. Our curriculum is functional-first, using abstraction, functional decomposition, immutable data, and powerful higher-order functions such as map, keep, and combine. Based on a decade of experience teaching thousands of first-time programmers, we believe starting with functions teaches students the habit of solving problems functionally when possible, helping to avoid bugs in later programming experiences.

This paper presents PlushPal, a web-based design tool for children to make plush toys interactive with machine learning (ML). With PlushPal, children attach micro:bit hardware to stuffed animals, design custom gestures for their toy, and build gesture-recognition ML models to trigger their own sounds. We describe how, in the context of storytelling, PlushPal introduces core concepts in ML including data sampling and model evaluation. We conducted online workshops and in-person play sessions with 11 children between 8-14 years old building interactive stuffed animals with PlushPal. In these play sessions, we observed how children imagined bringing their toys to life using ML, as well as how children’s data literacy changed as a result of experimenting with sensors, data sampling, and building their own ML models. Our work contributes a novel design space for children to express their ideas using gesture, as well as a description of observed debugging practices, building on efforts to support children using ML to enhance creative play.