In summary: Researches upload valid PostScript programs that capture all pages printed indefinitely in the future, and provide a facility to download them. They then show how this can be exploited by any website via some clever tricks with WebRTC and CORS.