The American Standard Code for Information Interchange is a 7-bit character encoding. Work on the standard began in 1960 and the first edition was published in 1963. The standard attempted to include characters which were common in other character encodings (such as EBCDIC) and equipment (e.g. typewriters, printers, teletypes) at the time.

Which characters/symbols/glyphs which were in use by other encodings/equipment in 1963 or earlier didn't (eventually) make it into ASCII?

Because ASCII is an "American" code, non-English letters don't count. Nor do things that were added in later versions of ASCII (e.g. curly brackets).

Examples: The cent sign ¢ was common on many typewriters and was bit pattern 100111 in the FIELDATA character set used by UNIVAC and Unisys.

The square lozenge ⌑ was part of the first BCDIC character set in 1932, and continued to be used in its successor EBCDIC. It was also used in the IBM 026 punched card code.

  • The request is way to broad to be answered in any useful fashion. There where hundrets of code sets already in use for various applications from teletypes to punch card machines.
    – Raffzahn
    Commented Feb 9, 2019 at 18:55
  • @Raffzahn: The question isn't about codesets; it's about characters.
    – DrSheldon
    Commented Feb 9, 2019 at 18:56
  • 5
    The Flexowriter was fairly common where I came from, and it had a good set of characters for representing Algol 60. These include multiply, integer divide, logical and, logical or, logical not, etc. I exclude characters produced by overstriking, since that's a function of equipment, not coding (Flexo underscore conveniently did not move the print position). Here's a partial listing.
    – dave
    Commented Feb 9, 2019 at 19:43
  • 1
    If you include conventional "printing equipment", pretty much any glyph in any international language. Plus a large set of math symbols, etc. Plus hundreds of ligatures, for some alphabets (e.g. Greek). The question is neither narrow nor clear IMO.
    – alephzero
    Commented Feb 9, 2019 at 20:31
  • 1
    Underbar also has the nice advantage of functioning as an underline. Plenty of programs used "print twice" for bold and "print underbar" for underline before dot-matrix printers & laser printers provided a way to print everything at once. Commented Feb 10, 2019 at 17:56

4 Answers 4


Small letters. These were included in the 1967 ASCII standard but not the 1963 one.


The account, amount and routing characters from the MICR sets: CMC-7 (Groupe Bull, 1957) and E13B (ABA, 1958-59).

1963 ASCII also used to have and where we now have ^ and _


Some symbols in the CDC display code, while assigned ASCII counterparts, haven't gotten into ASCII by themselves, nor into the "near" ASCII extensions, like ¢ or ¤ (that's U+00A4 CURRENCY SIGN, not U+2311 SQUARE LOZENGE). For example, the ≡ character, code 48 in the CDC set, is now U+2261 IDENTICAL TO.


The cents symbol was not in ASCII, was in IBM's EBCDIC, and therefore in keypunches, as mentioned the 026 and subsequent 029 and if I recall right the more automated 129

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .