The Innovators

Isaacson's organising argument is that the popular model of invention — a lone genius with a flash of insight — describes almost nothing in this history, and that the actual pattern is collaboration between people with complementary capabilities.

His recurring pairing is the visionary and the engineer, and his claim is that neither produces anything alone. The book is structured to demonstrate this repeatedly rather than to argue it once.

The lone-genius model describes almost nothing here. Every chapter is a collaboration between complementary people.

Devmark’s reading — not a quotation from the book

He opens with Lovelace and her notes on Babbage's Analytical Engine, crediting her with the insight that such a machine could operate on symbols rather than only numbers — that it could be general rather than arithmetic.

The extent of her contribution is debated among historians, and Isaacson takes a favourable view without much engaging the dispute. The framing choice is deliberate and sets the book's interest in overlooked contributors.

His middle chapters cover Turing's theoretical work, the wartime code-breaking effort, ENIAC, and the transistor at Bell Labs.

His treatment of the ENIAC programmers — six women whose work was substantially uncredited at the time — is one of the book's genuine contributions to general readership, restoring a history that had been almost entirely written out.

His account of Bell Labs as an institution is the most useful organisational material: a deliberate mixing of theorists and practitioners, with architecture designed to force encounters between people working on different problems.

The subsequent chapters on Shockley, Fairchild and the founding of Intel trace how the institutional model moved west and mutated into something faster and more commercial.

His two parallel strands: the network, developed through government-funded research with an unusually collaborative and non-proprietary culture, and the personal computer, developed largely through commercial competition.

His argument that the internet's architecture reflects the values of the institutions that built it — distributed, permissionless, resistant to central control — is the book's best point about how organisation shapes artefact.

The network's architecture reflects the values of the institutions that built it, which is the book's sharpest point.

Devmark’s reading — not a quotation from the book

The later chapters cover the personal computer software industry, the creation of the web at CERN, and the search engines that eventually organised it into something navigable.

These are the most compressed and the least original sections of the book. The material is thoroughly covered elsewhere, often by participants, and Isaacson is summarising received accounts rather than contributing anything new. A reader already familiar with this period can move through them quickly without missing the argument.

A thread running throughout: the recurring choice between building machines that replace human judgement and machines that augment it.

His sympathy is clearly with the augmentation tradition — Licklider, Engelbart — and he treats the artificial intelligence project as the road less taken, which reads differently now than when the book was published.

Isaacson is a synthesiser rather than an original researcher, and specialists in each of the periods he covers have noted simplifications, compressions and occasional factual errors in his account.

The collaboration thesis is also applied more uniformly than the history supports. Some contributions genuinely were substantially individual, and the book's framing occasionally flattens those cases to fit the pattern it has committed to demonstrating.

Briskly, and it covers an enormous span without losing the thread. Isaacson is a professional biographer and the pen portraits are efficient and memorable.

Its length is the main obstacle — it is long, and the later chapters cover ground most readers already know from elsewhere.

The collaboration thesis, which is genuinely useful as a corrective: when you encounter a story about a single inventor, ask who else was in the room, because there was almost always someone.

And the institutional argument — that Bell Labs, the ARPA-funded labs and the Homebrew Club produced what they did partly because of how they were organised, which is a design question rather than a matter of luck.