At M.I.T., Jacobson and a group of undergraduates made lists of requirements, methods, and materials. One of their tenets was: RadioPaper must reflect, like real paper. It must not emit. It couldn’t be based on some improved type of liquid-crystal screen, no matter how high its resolution, no matter how perfectly jewel-like its colors, no matter how imperceptibly quick its flicker, because liquid crystals are backlit, and backlighting, they believed, is intrinsically bad because it’s hard on the eyes. RadioPaper also had to be flexible, they thought, and it had to persist until recycled in situ. It should hold its image even when it drew no current, just as paper could. How to do that? One student came up with the idea of a quilt of tiny white balls in colored dye. To make the letter “A,” say, microsquirts of electricity would grab some of the microballs and pull them down in their capsule, drowning them in the dye and making that capsule and neighboring capsules go dark and stay dark until some more electricity flowed through in a second or a day or a week. This was the magic of electrophoresis.