Digital Communication - from Morse to Unicode
| System | Year | Encoding of "A" | Bits |
|---|---|---|---|
| Morse Code | 1844 | · − | Variable-length |
| Baudot Code | 1870s | 00011 | 5 bits |
| ASCII | 1963 | 01000001 | 7 bits |
| UTF-8 | 1991 | 01000001 | 8 bits (same as ASCII for Latin letters) |
This morse code table shows all letters, digits and some punctuation. Much more than that is not relevant to the simple equipment used.
Morse code is send over a telegraph line using a key on one end, and a sounder on the other. A key is basically just a switch with a spring, a sounder is an electromagnet with a metal bar that makes a clicking sound.
Baudot invented a device that looked like a 5-key piano. The timing of pressing the keys had to be exactly right, a clicking sound would let the user know when to release the keys. The speed at which codes could be entered would become known as Baud Rate.
Notice that the code itself is stateful. By default, you would be entering letters, but with the figures command, subsequent code could represent a digit or punctuation. These type of instruction codes are known as control code. Another example would be the bell code, that actually rang a bell.
Codes were written to a paper tape, later devices could also use paper tape as input and have Baud Rates much higher than humans could type. Typewriter-like devices could write text on a paper based on the paper tape. Even later, all this functionality would be combined in a single device called a Teletype.
With the introduction of the 7-bit ASCII standard, now there was room for lowercase letters and more special characters and control codes. Ascii was the reason that the 8-bit byte became the norm for not just data communication, but computer architecture in general.
The Ascii control codes were used both for instructing teletypes such as shown here, as well as other types of hardware terminals. Control codes such as Start of Heading (SOH), Start Of Text (STX), Acknowledge (ACK), etc were used to structure messages send over a network.
With the globalization of the internet, a universal character set was
needed to represent all possible languages and catface emoji's. 😻
The 64-bit character set is continually growing, which proves the need to have
encodings like UTF-8 to prevent a random text to have 4x the size of Ascii
text.
UTF-8 is a superset of Ascii, so all code based on Ascii would stil work as expected. That is, except for counting the number of characters, because 1 character can now suddenly be multiple bytes.