Matthew Raupp, an associate professor of UX design at the College for Creative Studies in Detroit and a member of Creative Coding Detroit, ran a hands-on workshop with Collage Machine, a Processing tool he has developed over several years for his art practice and his classes. It aims to balance artistic control and chance, and needs no coding experience. His path to creative code ran from a Commodore VIC-20 through Macromedia Director, Flash and ActionScript to Processing and p5.js.
Collage Machine began with early sketches that loaded a random photo on each click, then placed images randomly within the canvas, then made them draggable. The current version is the 36th. It began fully hand-coded, and in later versions he worked with ChatGPT, Gemini and Claude in a back-and-forth where he stays the author, asks the assistant to explain anything he does not understand, and builds in small increments. Conceptually he draws on Dada and Fluxus, which embraced chance, instructions and found material, and on contemporary collage artists and designers who build their own tools. He wants students to be tool builders, not only tool users.
The tool pulls from three folders of public domain material he has collected: objects, word fragments and graphics. Each button adds a random item at a random position and size. Participants then drag, scale, rotate, duplicate and delete, and can add colored shapes that blend in multiply mode. A dice button adds one of each type plus a color. His favorite controls randomly reposition and rescale selected pieces, which can turn a carefully arranged composition into a fresh starting point. There is no undo. Students ask for it, but he said it is hard to build and asks them to embrace what they are given. Several participants agreed with that choice, and many shared their compositions in chat as the session went on.
Raupp then thought aloud through a composition the way he does in class: placing a focal object on the rule of thirds, clustering and duplicating, finding shared shapes and sight lines, pushing scale to extremes, letting elements bleed off the edge, activating negative space and reading for hierarchy. In his sophomore design semiotics course, students load their own images, after first working in Photoshop, Illustrator and paper collage. He uses the tool to discuss how every tool shapes its outcomes, and how a library of images reflects its collector’s lens. Asked about authorship and loss of control, he said the class reads and discusses both, and that he hopes to let the tool pull images from online sources in the future.
Three takeaways
- Build tools, not just use them. A custom tool lets an artist set the constraints that shape their work, and makes visible how every tool shapes outcomes (00:20:22–00:21:40, 00:57:00–00:58:35).
- Mix control and chance. Random placement, the dice and reposition-and-rescale controls offer new starting points that the artist can then refine (00:35:34–00:39:30).
- Name the composition strategies. Rule of thirds, clustering, duplication, scale, bleed, focal point and hierarchy give students a vocabulary for what they are seeing (00:43:33–00:49:37).
Try it
From the workshop: download Collage Machine, replace the objects folder with ten of your own photos, and make three compositions. For each, press P and S several times and save one variation you would not have made by hand. Write one sentence on what the randomness changed.
Watch and explore
Replay: Watch on YouTube
- Processing download
- Collage Machine sketch (ZIP download)
- Matthew Raupp on Instagram
- Wiki Spy, shared by a participant as a related idea for pulling images from the web.
From the original program
Published before the event
Collage Machine, an Algorithmic Approach to Play
Join designer and educator Matthew Raupp for a hands-on exploration of Collage Machine, a Processing tool for digital collaging that strikes a balance between artist control and chance. Matthew will share the evolution of the tool and how he integrates it into both his studio practice and classroom. Attendees will make a series of compositions using interactive controls to manually position, resize, and rotate elements, paired with randomizers that continuously generate playful, unexpected results.
About the presenters
Matthew Raupp
He/Him/His
Matthew Raupp is a Communication Design educator with over 25 years of teaching experience and a former partner of Defrost Design. His practice centers on "serious play," using code not just as a technical skill or means of production, but as a sandbox for experimentation. In an evolving technological landscape, Matthew views coding as a way to transition from a being a tool user to a tool maker, a philosophy that comes to life in his latest creative coding endeavor, Collage Machine, an algorithmic tool that strikes a balance between artist control and chance. By bringing these generative systems into both studio practice and the classroom, he helps designers shift beyond closed software constraints, using code to sketch, generate unexpected outcomes, and invent their own ways of making.