Strictly speaking, it's Lempel-Ziv coding. LZ compresses text by replacing the second occurrence of a phrase with a pointer to the first, as above, whereas Huffman represents more frequently used characters with shorter bit strings, like Morse code.