I am trying to trace (as far as known) the first example of a digital computing device. I am discounting (no pun) an abacus because I regard that as a device for representing, storing and manipulating discrete, not binary digital, data values.

I have read that in 1937 Church and Turing published theoretical papers on computing. My understanding is that the (universal?) Turing machine does not explicitly require a digital representation of data and operations (maybe that is wrong).

I am trying to track down the point that it became clear that both data and operations could be represented by digital codes (maybe it was Church-Turing?). I have read that the earliest binary digital computers were produced by

1938: Konrad Zuse: Konrad Zuse creates the Z1 Computer a binary digital computer using punch tape

1939 George Stibitz: George Stibitz develops the Complex Number Calculator - a foundation for digital computers

Was this the time when the relationship between high-level operations (such as integer multiplication) and manipulating binary digits became explicit?

  • $\begingroup$ Consider an automated circuit that can transcribe morse code into letters. It is binary, but is it a device? $\endgroup$ Commented Mar 28, 2017 at 3:44
  • $\begingroup$ It is hard to tell what you are asking. Sorry, but I do not follow your distinction between "discrete" and "digital" devices. You also mix two separate issues, "binary/digital computing devices" and "data and operations represented by digital codes". The latter is much more recent, Ada Lovelace is often credited with writing a first computer program in 1840-s. But systematic development of the idea only appears in von Neumann's 1945 draft. $\endgroup$
    – Conifold
    Commented Mar 28, 2017 at 4:23
  • $\begingroup$ binary is just a notation. not essential, any other would do as well. in principle you can treat any discrete state machine as a "digital" computing device, and such machines were available in ancient times. i think. ;) $\endgroup$
    – mobileink
    Commented Mar 28, 2017 at 20:48
  • $\begingroup$ "...Turing machine does not explicitly require a digital representation of data and operations..." I don't recall but i doubt Turing said anything about "digital" anything. Strictly speaking TMs do not involve signs or representations. They just transform shapes according to rules. Whether those shapes have meaning is a separate queation. is your main question about "data and operations ... represented by digital codes"? but traditional math already did this. maybe you mean sth like "elimination of the distinction between operations and data"? $\endgroup$
    – mobileink
    Commented Mar 28, 2017 at 21:03

2 Answers 2


Digital devices were quite common since 19th century, they were called cash registers:


Many shops were equipped with them. Similar device for scientific/engineering purposes was called arithmometer,


In 1970s I was solving PDE's with this arithmometer:


or adding machine. The earliest ones appear in 17th century (Pascal is credited with invention of one.)

Remark. The main feature of modern computers which sets them apart is that they are programmable, not that they are digital. But programmable computers also existed in 19th century (Babbage machine, for example).


There were also quite sophisticated analog computers (as opposite to digital) and they were also in use since 19th century.


It has been claimed that the Jaquard Loom (1804) was a digital computer. (But not electronic, of course.) We could even say that it was binary, since on the punched cards that control it, in each locaton on the card there was either a hole or not.

Then there were music boxes driven by clockwork, where the notes where controlled by holes punched in a metal disk.

music box


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.