Decoding the Heavens

Jo Marchant

The bronze fragments of an ancient Greek device have puzzled scholars for more than a century after they were recovered from the bottom of the Mediterranean Sea, where they had lain since about 80 B. C. Now, using advanced imaging technology, scientists have solved the mystery of its intricate workings. Unmatched in complexity for a thousand years, the mechanism functioned as the world’s first analog computer, calculating the movements of the sun, moon, and planets through the zodiac. In Decoding the Heavens, Jo Marchant details for the first time the hundred-year quest to decode this ancient computer. Along the way she unearths a diverse cast of remarkable characters-ranging from Archimedes to Jacques Cousteau-and explores the deep roots of modern technology, not only in ancient Greece, but in the Islamic world and medieval Europe. At its heart, this is an epic adventure story, a book that challenges our assumptions about technology development through the ages while giving us fresh insights into history itself.

ScienceScience · high confidenceScience81 signalsLlm Classifier: Science -> ScienceOpen Library: Science -> ScienceOpen Library: History -> HistoryOceans, Rivers & WaterScience & DiscoveryTechnology, Computing & AIEurope & Russia
Experimental book profileGenerated by GPT-5.4 nano · may contain inaccuracies

This is an unverified interpretation of the catalog description, offered as an opt-in discovery experiment—not as bibliographic fact.

  1. 01 Central figures

    None extracted with sufficient confidence.

  2. 02 Central places

    None extracted with sufficient confidence.

    These suggestions did not meet the normal display threshold and are more likely to be wrong.

    • Mediterranean Sea55%
  3. 03 Suggested argument

    No argument inferred with sufficient confidence.

  4. 04 Reading orientation

    38 / 100 academic

    Serious trade

    An estimate of intended readership and scholarly apparatus—not quality or importance.

Confidence percentages are the model's own estimates. Profile confidence: 55%.