First Draft of a Report on the EDVAC

First Draft of a Report on the EDVAC. This seminal document, circulated in June 1945, is widely considered the foundational blueprint for the modern stored-program computer. Authored by mathematician John von Neumann, it synthesized ideas from the ongoing EDVAC project at the Moore School of Electrical Engineering into a coherent logical design. Its description of a unified architecture, where instructions and data reside in a single memory, fundamentally shaped the development of subsequent computers worldwide.

Background and authorship

The report emerged from collaborative work on the EDVAC project, the successor to the pioneering ENIAC at the University of Pennsylvania. Key figures like J. Presper Eckert, John Mauchly, and Herman Goldstine contributed to the underlying concepts during discussions at the Moore School of Electrical Engineering. John von Neumann, a consultant on the Manhattan Project and a member of the Institute for Advanced Study, joined the group and compiled their ideas into this cohesive draft. His involvement followed earlier work with Stanislaw Ulam on computational methods and his exposure to the ENIAC team through Herman Goldstine at the Aberdeen Proving Ground.

Structure and content

The document is organized as a detailed technical specification, beginning with a critique of the ENIAC's limitations. It systematically outlines the components of a new machine, dedicating sections to the arithmetic organ, the central control, and a unified memory system. A significant portion details the proposed use of delay line memory, a technology explored during World War II in projects like Project Pigeon and radar systems. The report also introduces a formal notation for instruction codes and discusses the logical flow of operations, framing the computer as a sequential, automated device.

Key concepts and innovations

Its most revolutionary proposal was the stored-program concept, where both data and instructions are stored in the same addressable memory unit. This eliminated the need for manual rewiring, as required by the ENIAC, and allowed for conditional branching and self-modifying code. The report formally described a central processing unit with distinct arithmetic and control units, an architecture later dubbed the von Neumann architecture. It also emphasized the use of binary over decimal systems, influenced by Claude Shannon's work at Bell Labs, and detailed the serial execution of instructions fetched from memory.

Influence and historical significance

The draft's distribution had an immediate and profound impact on postwar computer development globally. It directly inspired the design of the IAS machine at the Institute for Advanced Study, the Manchester Baby at the University of Manchester, and EDSAC at the University of Cambridge. The logical framework was adopted for early commercial systems like the UNIVAC I and influenced architects such as Maurice Wilkes and Alan Turing. This document effectively established the paradigm for first-generation computers, shaping projects from the Whirlwind at MIT to early Soviet efforts like the MESM.

Publication and distribution

The report was never formally published as a book or journal article during its era. Instead, it was mimeographed and distributed on June 30, 1945, by Herman Goldstine, who listed only John von Neumann as author. This limited circulation, primarily to about two dozen individuals and institutions like the National Defense Research Committee and the Radiation Laboratory, nevertheless ensured its rapid dissemination within the nascent computing community. Its status as an internal document later fueled controversies over credit, particularly with J. Presper Eckert and John Mauchly, and it was eventually reprinted in later collections on the history of computing.

Category:Computer science papers Category:1945 documents