The genially unorthodox author of Mindstorms (1983) again makes a stimulating case for computers as a primary route to knowledge, revising and expanding earlier observations in view of disappointing school policies of the past dozen years. Rejecting most schools as ``sluggish and timid'' in assuring access to learning, Papert (Mathematics and Education/MIT) divides the conservative education world into ``Schoolers'' (who acknowledge underlying problems but focus on short-term urgent ones) and ``Yearners'' (who create their own small-scale alternatives) as he considers why technology hasn't revolutionized school learning. Championing computers for offering forms of learning that can be ``quick, immensely compelling, and rewarding,'' Papert contends that Logo (the computer language he conceived) is a superior mode of learning for young children, closer to their informal learning style than traditional classroom approaches and invaluable as a medium for most areas of study. But schools have ignored computers' broad capacities, he finds, isolating these tools from the learning process instead of integrating them into all areas of instruction. Papert offers a steady supply of examples--from his own extensive experience as well as from assorted classrooms--providing evidence of computers as powerful learning allies. He also understands the nature of learning--the importance of the personal element in any classroom exchange; the need to adapt a learning-environment design to its social and cultural milieu; the ``internal censors'' that students bring to required work; and the way that ordered ideas can emerge from an imprecise, undirected process. Even those who resist Papert's belief that the foundation of modern schooling is faulty will agree with his central theme that the ability to learn new skills is the most critical skill of all- -and that computers have a unique, accelerating role to play in developing that ability.