Manchester Baby

The Manchester Baby — officially the Small-Scale Experimental Machine (SSEM) — was a computer built at the University of Manchester that ran the world's first stored program on 21 June 1948. It was never meant to be a production machine: it existed to prove that the new Williams tube memory worked. It proved far more — it proved that a computer could hold its own program.

Why it existed

Memory was the hard problem of 1940s computing. Alan Turing and others had designs on paper, but no practical technology could reliably store thousands of bits. Frederic Calland Williams and Tom Kilburn at Manchester had developed the Williams tube — a cathode-ray tube that stored binary data as persistent charge patterns on the phosphor faceplate, readable and rewritable at electronic speed. To test it, they built the simplest possible computer around it: the Baby.

The machine

The Baby was tiny by any era's standard — a single 32-word memory of 32-bit words, a small accumulator, and a control unit cycling through instructions at roughly 700 per second. It ran just seven instructions — enough to load, store, subtract, skip, jump, and halt. Its cabinet filled a room anyway; by modern measure it was room-sized, CRT-glowing, and blink-and-you-miss-it fast.

The historic run, 21 June 1948: a 17-instruction program (Kilburn's, calculating the highest factor of 2¹⁸) executed successfully — the first time a stored program had ever run anywhere.

What it proved

• Stored program worked in practice — the EDVAC idea left the page and ran.
• The Williams tube worked — it became the dominant memory technology of the early 1950s, used in the IBM 701, Whirlwind, and others.
• Manchester won the race — the Baby beat EDSAC by a year, though EDSAC ran the first full, usable stored-program service.

Within months the Baby was rebuilt and expanded into the Manchester Mark 1 (1949), which fed directly into the Ferranti Mark 1 (1951) — the world's first commercial general-purpose computer. Alan Turing joined Manchester in 1948 and wrote early programs for the Mark 1, including routines for playing chess — which the machine was still too weak to finish.

See also

• EDSAC
• EDVAC report
• Alan Turing
• John von Neumann
• History of Programming

References

1. Lavington, Simon. Early British Computers (Manchester UP, 1980).
2. Wilkes, Maurice V. Memoirs of a Computer Pioneer (MIT Press, 1985).
3. Lavington, Simon. A History of Manchester Computers (NCC Publications, 1975; 2nd ed. 1998).