LITERARY THEORY FOR ROBOTS

An intriguing glimpse into how the secret machinery that makes our technology work has deep roots in philosophy, poetry, and linguistics.

We’ve become so used to computers that can understand what we put on a screen—checking our spelling and correcting our grammar—that we forget that they use complex, clever processes to do so. In this brief, pithy book, Yi Tenen, a former software engineer at Microsoft and current affiliate at Columbia University’s Data Science Institute, suggests that giving computers literacy should be seen as one of the most essential technological feats of the 20th century. Locating the beginning of the story is difficult, since Arabic philosophy, Chinese numerology, and other intellectual traditions from around the world and across centuries all have antecedents. Mechanical cypher systems also played a role, as did early attempts at creating a math-based language. Babbage, Bacon, and Leibniz were all interested in giving machines an interactive element, and Turing made a huge contribution with his theories of coding. For decades, people had contrasting ideas on how to get computers to “read,” but all ran into the problems of defining intelligence, communication, and understanding. As Yi Tenen shows, we tend to see the human mind as a metaphor, but the author believes that this is not really an appropriate comparison: The repetitive, algorithmic pattern of machine learning does not reflect the conceptual, intuitive nature of human development. The advent of personal computers started tech on the path to artificial intelligence, although getting everything to work together requires extensive collaboration and creativity. Yi Tenen, stirring some wit and anecdotes into the story, sets out the material in non-technical terms, making for an entertaining, informative read.

An eclectic and erudite tale of how wide-eyed visions become smart, interactive tools.