Library · paper

The Early History of Smalltalk

Alan Kay
1993·ACM SIGPLAN Notices

Source: https://dl.acm.org/doi/10.1145/155360.155364

The other half of the software history that Brooks and the Unix tradition represent.

Kay and the Xerox PARC team invented objects, GUIs, and the idea that computing should be a medium for human expression — not a tool for data processing.

For product people: the argument that the interface is not a layer on top of the system but the system itself.

Central argument

Kay argues that Smalltalk emerged from a core insight he calls the 'Dynabook vision': that the computer should be a dynamic medium for creative thought, not a data-processing machine. The central thesis is that object-oriented programming and the GUI were not technical inventions bolted onto existing paradigms but the necessary consequences of treating the interface and the system as a unified expressive environment. Kay traces how PARC's work on message-passing objects, windows, and direct manipulation all derived from this philosophical commitment — that the user's mental model and the system's architecture should be the same thing.

Critique

Kay's account is a founder's narrative, and it carries the distortions of that genre: competitors and parallel inventors (Simula, Sketchpad's successors, the Alto hardware team) are acknowledged briefly but the intellectual lineage is shaped to foreground his own conceptual contributions. More substantively, the claim that 'the interface is the system' was easier to sustain at PARC, where the same people designed hardware, OS, language, and application — a condition almost no product organisation can replicate, which leaves the central thesis inspiring but practically underspecified.

Why it matters for product

For a CPO, Kay's argument challenges the standard org chart assumption that design sits on top of engineering: if the interface is not a layer but the architecture itself, then separating UX from systems decisions at the team structure level produces incoherent products by construction. More concretely, it reframes discovery work — the question is not 'what do users want to do' but 'what kind of thinking does this medium make possible', which is a harder and more generative brief for product teams operating in platform or tool-building contexts.