How did people realize they could do logic with electronics? Are there anecdotes or records of the first realizations? I'm wondering about the first "eureka" moments.
From the Wikipedia article, Boolean algebra:
In the 1930s, while studying switching circuits, Claude Shannon observed that one could also apply the rules of Boole's algebra in this setting, and he introduced switching algebra as a way to analyze and design circuits by algebraic means in terms of logic gates. Shannon already had at his disposal the abstract mathematical apparatus, thus he cast his switching algebra as the two-element Boolean algebra.
The article on Claude Shannon gives some more detail:
In 1936, Shannon began his graduate studies in electrical engineering at MIT, where he worked on Vannevar Bush's differential analyzer, an early analog computer. While studying the complicated ad hoc circuits of this analyzer, Shannon designed switching circuits based on Boole's concepts. In 1937, he wrote his master's degree thesis, A Symbolic Analysis of Relay and Switching Circuits, A paper from this thesis was published in 1938. In this work, Shannon proved that his switching circuits could be used to simplify the arrangement of the electromechanical relays that were used then in telephone call routing switches. Next, he expanded this concept, proving that these circuits could solve all problems that Boolean algebra could solve. In the last chapter, he presents diagrams of several circuits, including a 4-bit full adder.
Using this property of electrical switches to implement logic is the fundamental concept that underlies all electronic digital computers. Shannon's work became the foundation of digital circuit design, as it became widely known in the electrical engineering community during and after World War II. The theoretical rigor of Shannon's work superseded the ad hoc methods that had prevailed previously. Howard Gardner called Shannon's thesis "possibly the most important, and also the most noted, master's thesis of the century."
As with so many other important developments in logic and computer science, it was almost certainly the mathematician and philosopher Charles Sanders Peirce, whose work predated Shannon's by decades:
Of course, it is a manifestation of genius to have an idea long before it is understood and appreciated. Let me close by outlining the background for another of Peirce's logical ideas of great originality, the idea for a general-purpose relay computer, which was fifty years ahead of its time. The sequence of events is as follows:
- Peirce stimulated Alan Marquand to invent and build a mechanical logic machine superior to that of William Stanley Jevons. This machine is described in Peirce's Logical machines, vol. III, pt. 1, pp. 625–632.
- This machine was built in the early 1880s. At about the same time, Peirce conceived the sufficiency of "not-and" and "not-or," together with the use of a truth-table as a decision procedure for tautologyhood.
- In a letter to Marquand dated 1886 Peirce suggested the use of relays for Marquand's machine and showed how to achieve "and" and "or" with relays. "... it is by no means hopeless ... to make a machine for really very difficult mathematical problems (ibid., p. 632).
- Marquand then prepared a wiring diagram for a relay version of his mechanical logic machine.
(Source: Arthur W. Burks, [“The New Elements of Mathematics” (book review) p. 917, Bulletin of the American Mathematical Society, vol 84, issue 5 (September 1978). Boldface emphasis is mine.)
Quoting from Peirce's 1886 letter to Marquand:
… it is by no means hopeless to expect to make a machine for really very difficult mathematical problems. But you would have to proceed step by step. I think electricity would be the best thing to rely on. Let A, B, C be three keys or other points there the circuit may be open or closed. As in Fig 1, there is a circuit only if all are closed; in Fig. 2 there is a circuit if any one is closed. This is like [logical and & logical or] in Logic.
(Source: Writings of Charles S. Peirce: A Chronological Edition, vol. 5 (1884–1886) p. 422. Indiana University Press, 1993. Christian J. W. Kloesel et al., editors.
Peirce was an amazing case of someone who was so far ahead of his time that his work couldn't be appreciated by his contemporaries. He was mostly ignored in his lifetime, but he managed to anticipate a huge number of important logical and mathematical developments that then had to be rediscovered much later. For example, he invented lattice theory in the 19th century, but nobody really paid attention until Garrett Birkhoff reinvented it in 1935. Point 2 in the Burks quotation above observes that Peirce invented NAND logic (still the basic logic of microchips today) but the credit is usually given to Henry Sheffer who discovered it 23 years later. Stanford Encyclopedia of Philosophy article about Peirce.
As far as "eureka" moments go, I think the application of Boolean logic to electronics became inevitable the moment Boolean Algebra was formalized by George Boole in
The Mathematical Analysis of Logic in 1847. Wikipedia
It could also be argued that this "eureka" occurred a decade prior to the formalization of Boolean logic when Charles Babbage attempted the construction of his Analytical Engine in 1837, a device containing
an arithmetic logic unit, control flow in the form of conditional branching and loops, and integrated memory.
The argument here is strong if one considers that, from a computational perspective, both mechanical and electronic logic gates are equivalent. The replacement of mechanical components with cheaper, more reliable electronic ones was not limited to logical components and was widespread through all industries. Had Babbage had the basic electronic components available, one can imagine he would have utilized them for this sort of logic in exactly the same way he did mechanical ones.
A third possible "eureka" could be the meeting of Babbage and Boole at the Great London Exposition in 1862:
The two are said to have discussed this "thinking engine," which Babbage never completed. But it became a building block for modern computing.
Yet another "eureka" milestone could be the realization of Babbage's Analytical Engine dream with the completion of Howard Aiken's functioning, electomagnetic Automatic Sequence Controlled Calculator at Harvard in 1937.
Lastly, we can certainly peg the moment no later than (as mentioned in @the-photon's answer) in Claude Shannon's formalization of the mairrage of Boolean Logic with electronic components at MIT in 1938.
This excellent Atlantic article answers your question at length. Here's the closest thing to a Eureka moment:
Today, Boole’s name is well known to computer scientists (many programming languages have a basic data type called a Boolean), but in 1938 he was rarely read outside of philosophy departments. Shannon himself encountered Boole’s work in an undergraduate philosophy class. “It just happened that no one else was familiar with both fields at the same time,” he commented later.
Strowger's 1889 automatic telephone exchange was certainly a practical and real world use of digital logic via electromechanical means. Solving other pulse/state logic problems with relays and other electromechanical parts cannot have been an entirely new concept at the latest after this point in time.
Combining the facts "relays are slow and noisy" and "gas discharge and/or vacuum tubes and their technical successors are faster and can do the same job" to "let's use literal electronics for digital logic" appears almost trivial.
Some added explanations: "Gas discharge Tubes" as in Thyratrons, or even plain neon lamps (these have a strong hysteresis between striking and extinguishing voltages and can thus act as a memory element), or more complex thyratron-derived devices like dekatron counting tubes. Earlier production design vacuum tubes (up into the 1940s - the ENIAC design used that generation and had severe problems with it :) actually hated being used as hard on/off switching elements (being left with full voltage applied but switched hard off a lot progressively damaged the cathode coating. keyword is "cathode interface", or "zwischenschichtbildung" in german literature*); vacuum tubes that were reliable in that function were introduced for 50s/60s era industrial control equipment...
*Mentioning that because datasheets might only exist in English,German,Dutch or French for some of these types...